R-loops, three-stranded DNA-RNA hybrids, are formed by invasion of an RNA strand into duplex DNA and can be assembled either co-transcriptionally (cis R-loops) or by local unwinding of DNA and subsequent hybridization of homologous RNA fragments (trans R-loops). R-loops have long been regarded as unwanted structures, potentially leading to transcription or replication fork collisions as well as leading to a single DNA strand, which is energetically less stable than double stranded DNA and therefore prone to damaging agents. Recent advances have, however, challenged this view and shown a subclass of R-loops to be involved in epigenetic regulation of a number of genes, leading to R-loops as being often described as a double-edged sword: A threat to genome integrity (unscheduled R-loops) and epigenetic regulators of gene expression (regulatory R-loops).
Regulatory RNAs play key roles in development and disease and yet we are still at the beginning of understanding their biology and mode of action. Among regulatory RNAs, R-loops, prevalent DNA:RNA hybrids, which lie at the interface of different nuclear processes and neurodegenerative disease, have come into the limelight as novel gene regulators. Accumulating evidence indicates that lncRNAs can act as guide RNAs to target gene regulators to genes via R-loops. This is a breakthrough in understanding how these DNA:RNA hybrids act as signals and sets the stage to address fundamental questions regarding the epigenetic mechanisms of R-loop function and their biology:
• What is the function of regulatory R-loops?
• Do they form via cis- or trans acting RNAs?
• How are the potentially detrimental effects of regulatory R-loops constrained?
• How are regulatory R-loops regulated?
• What molecular determinants distinguish regulatory R-loops from unscheduled R-loops?
• How is the information of regulatory R-loops coded and decoded?
The objective of this project is to provide answers to these questions by making use of an epigenetic R-loop reader protein, Gadd45a, which we have identified previously.
The growth arrest and DNA damage inducible 45 (GADD45) family of proteins has three members: GADD45a, GADD45b, GADD45g. These small proteins (~18 kDa) share high sequence similarity and function in regulating stress response, DNA damage repair and cell cycle control. GADD45 proteins lack enzymatic activity and carry out their functions by recruiting enzymatic effectors. Intriguingly, recent work from our lab revealed GADD45a to function as an R-loop reader. The study showed that the TARID lncRNA (TCF21 Antisense RNA Inducing Demethylation) binds to the TCF21 gene promoter forming an R-loop. GADD45a binds this R-loop and in turn recruits TET1 and TDG to bring about active DNA demethylation and activate the TCF21 gene transcription. However, mechanistic insight into genome-wide GADD45a-mediated R-loop decoding is still missing.
Key tools have been obtained during the action, including a genome-edited Gadd45a gene harbouring an HA-epitope for facilitated Gadd45a detection as well as several Gadd45a,b,g triple mutant ESC lines.
A vast amount of data was obtained during the action, including genome-wide profilings to characterize the nuclear role of Gadd45. Only a minor fraction of the data has been published as yet. Comprehensive papers reporting these results will be released in the coming years. We anticipate that by the time of publication, we will have an understanding of the role of Gadd45 decoding regulatory R-loop hybrids in undifferentiated and differentiating mouse embryonic stem cells.
