Helicases in R-loop Formation and Resolution

Abstract

With the development and wide usage of CRISPR technology, the presence of R-loop structures, which consist of an RNA-DNA hybrid and a displaced single-strand (ss) DNA, has become well accepted. R-loop structures have been implicated in a variety of circumstances and play critical roles in the metabolism of nucleic acid and relevant biological processes, including transcription, DNA repair, and telomere maintenance. Helicases are enzymes that use an ATP-driven motor force to unwind double-strand (ds) DNA, dsRNA, or RNA-DNA hybrids. Additionally, certain helicases have strand-annealing activity. Thus, helicases possess unique positions for R-loop biogenesis: they utilize their strand-annealing activity to promote the hybridization of RNA to DNA, leading to the formation of R-loops; conversely, they utilize their unwinding activity to separate RNA-DNA hybrids and resolve R-loops. Indeed, numerous helicases such as senataxin (SETX), Aquarius (AQR), WRN, BLM, RTEL1, PIF1, FANCM, ATRX (alpha-thalassemia/mental retardation, X-linked), CasDinG, and several DEAD/H-box proteins are reported to resolve R-loops; while other helicases, such as Cas3 and UPF1, are reported to stimulate R-loop formation. Moreover, helicases like DDX1, DDX17, and DHX9 have been identified in both R-loop formation and resolution. In this review, we will summarize the latest understandings regarding the roles of helicases in R-loop metabolism. Additionally, we will highlight challenges associated with drug discovery in the context of targeting these R-loop helicases.

Keywords: CRISPR; R-loop; R-loop formation; R-loop resolution; RNA-DNA hybrid; cancer; disease; helicase.

Figure 1

Potential R-loops and RNA–DNA hybrids in cells.A, an R-loop forms during transcription, where a nascent RNA synthesized by RNA polymerase II can base pairs with a DNA template, forming an RNA–DNA hybrid and displaced non-template ssDNA. B, an R-loop forms at the telomere. Telomeres are transcribed into TERRA that binds to telomeric DNA, forming R-loop structures, leaving a displaced G-rich DNA strand that forms G4. C, an R-loop forms in CRISPR/Cas9. A guide RNA binds the target DNA strand while the nontarget strand is displaced, forming an R-loop structure, and Cas9 cleaves both strands. D, an RNA–DNA hybrid forms during DNA replication, where a primer RNA in a leading/lagging strand pairs with a DNA template. E, an RNA–DNA hybrid forms during DNA double-strand breaks (DSBs). A new RNA synthesized by RNA polymerase III at DSB pairs with a DNA template.

Figure 2

Alignment of senataxin orthologs and predicted human senataxin structure.A, the conserved helicase core domain is indicated in orange, and the seven helicase motifs and the accessory domains are indicated. The number of amino acids in each helicase is shown on the right. The exact boundaries of domains may differ. Most sequences are from Uniprot: human (Q7Z333), mouse (A2AKX3), rat (A0A8I5ZYQ6), Drosophila (B7Z0D7), zebrafish (A0A8N7UR12), baker’s yeast (Q00416), Arabidopsis (B6SFA4), and CasDinG (Q2FGY5); some are from databases: Wormbase (Eri-7, CE36108), Pombase (SPAC6G9.10c), and the frog is from Ref (156). B, predicted human senataxin structure by the Alphafold2. The conserved RecA1 and RecA2 domains, the structured N-terminal domain, the first (N) and last residue (C) are indicated. Amino acids are colored based on their per-residue confidence score (pLDDT), which values can be between 0 and 100. Low values indicate low confidence and high numbers indicate very confident predictions. Amino acids are either colored orange (pDLLT <50), yellow (70> pDLLT >50), light blue (50< pDLLT >70), or dark blue (pDLLT >90). Regions with pLDDT <50 may be unstructured in isolation. FHA, forkhead associated domain; NLS, nuclear localization signal.

Figure 3

Potential roles of helicases in R-loop metabolism. For R-loop formation helicases (greentriangle), they may utilize their helicase/translocase activity to remove any secondary structures or R loop suppressing proteins (blue oval) on the RNA to facilitate the invasion of RNA into dsDNA (e.g., UPF1), their strand annealing activity to facilitate RNA–DNA hybrid formation (e.g., Cas3) or their unwinding activity to unwind secondary structures on ssDNA(e.g., DDX1) or ssRNA (e.g., DHX9) and facilitate RNA–DNA hybrid formation. For R-loop resolution helicases (redtriangle), they may utilize their unwinding activity to separate RNA–DNA hybrids (e.g., SETX), utilize their strand annealing activity to anneal dsDNA and push away ssRNA (e.g., SMARCAL1), or separate dsDNA and lead to long ssDNAs that are degraded by nucleases (gray three-quarter circle) (e.g., CasDinG).

Figure 4

Phylogenetic analysis of R-loop helicases. The phylogenetic tree was constructed by the neighbor-joining method using the MEGA11 software with 1000 bootstrap replicates. The optimal tree is shown. Bootstrap values are shown at nodes. This analysis involved 26 amino acid sequences, 24 are from Homo sapiens, Cas3 from Streptococcus thermophilus, and CasDinG from Pseudomonas aeruginosa (Table 1).