The interplay of RNA:DNA hybrid structure and G-quadruplexes determines the outcome of R-loop-replisome collisions

Abstract

R-loops are a major source of genome instability associated with transcription-induced replication stress. However, how R-loops inherently impact replication fork progression is not understood. Here, we characterize R-loop-replisome collisions using a fully reconstituted eukaryotic DNA replication system. We find that RNA:DNA hybrids and G-quadruplexes at both co-directional and head-on R-loops can impact fork progression by inducing fork stalling, uncoupling of leading strand synthesis from replisome progression, and nascent strand gaps. RNase H1 and Pif1 suppress replication defects by resolving RNA:DNA hybrids and G-quadruplexes, respectively. We also identify an intrinsic capacity of replisomes to maintain fork progression at certain R-loops by unwinding RNA:DNA hybrids, repriming leading strand synthesis downstream of G-quadruplexes, or utilizing R-loop transcripts to prime leading strand restart during co-directional R-loop-replisome collisions. Collectively, the data demonstrates that the outcome of R-loop-replisome collisions is modulated by R-loop structure, providing a mechanistic basis for the distinction of deleterious from non-deleterious R-loops.

Keywords: DNA replication; G-quadruplex; Pif1; R-loop; RNase H; S. cerevisiae; biochemistry; chemical biology; chromosomes; gene expression; replisome; transcription-replication conflict.

Figure 1.. Preparation and characterization of R-loop-containing templates.

(A) Schematic of R-loop plasmid template. Plot shows G/C skew at Airn sequence. Graphic shows positions of potential G-quadruplexes composed of stacks of three G-quartets in non-template (top) and template (bottom) strand. 3× T: T7 terminator tandem repeat. (B) Schematic of co-directional (CD) and head-on (HO) R-loop-replisome collisions in experimental setup. Template strands: black; leading strand: red; lagging strand: blue; RNA: green. (C) Reaction scheme for preparation of R-loop-containing template. (D) Native agarose gel analysis of purified plasmid template. The gel was stained with ethidium bromide. (E) R-loop-containing template harboring ³²P-labeled RNA was mock-treated or digested with RNase H and analyzed by denaturing formaldehyde agarose gel-electrophoresis and autoradiography. (F) Electron microscopy (EM) analysis of R-loop templates. White arrows in center panels indicate S9.6-specific density. Thin and thick arrows in right panels indicate displaced non-template or RNA:DNA duplex, respectively. (G) Frequency distribution of R-loop distances from ClaI site in CD and HO orientation.

Figure 1—figure supplement 1.. R-loop template preparation and characterization by atomic force microscopy (AFM).

(A) Sequence of Airn sequence element. Top strand: non-template strand; bottom strand: template strand. QGRS Mapper (https://bioinformatics.ramapo.edu/QGRS/index.php) was used to identify sequences with G-quadruplex-forming potential in non-template and template strand, highlighted in orange and green, respectively. Numbers in brackets indicate G-quadruplex-forming potential. (B) pARS^R-loop was transcribed, the salt adjusted to 0.4 M NaCl, and either mock-treated or treated with RNase A. Reactions were de-proteinated, phenol/chloroform-extracted, filtered through Illustra MicroSpin G25 Spin column, and 2 μL of each sample analyzed by native agarose gel-electrophoresis and ethidium bromide staining. (C) S1000 gel filtration profiles of R-loop plasmid templates. RNA is ³²P-labeled. Fractions were analyzed by ethidium bromide staining (top) or autoradiography (bottom) after native agarose gel-electrophoresis. (D) Frequency distribution of location of S9.6-dependent electron-dense structures relative to ClaI site in CD orientation. (E) Representative AFM images of linearized R-loop templates. Templates were incubated with yeast RPA prior to deposition on mica.

Figure 2.. Both co-directional (CD) and head-on (HO) R-loops perturb normal fork progression.

(A) Schematic illustrating expected sizes of replication products. (B) Denaturing agarose gel analysis of replication products obtained on R-loop-free template. Left lead: Leftward leading strands; Right lead: Rightward leading strands. (C) Native (top) and denaturing (bottom) agarose gel analyses of replication products obtained on templates harboring Airn sequence in CD or HO orientation. Stall: Stalled rightward leading strands; Restart: Rightward leading strand restart product; Full-length: full-length rightward leading strand; RI: replication intermediates. (D) Schematic illustrating replication products observed in (C). (E) Two-dimensional gel analysis of replication products obtained in presence of R-loops (corresponding to lanes 4 and 8 in (C)). Products were digested with ClaI. (F) Replication products obtained in the absence or presence of RFC/PCNA or Pol δ.

Figure 2—figure supplement 1.. Both co-directional (CD) and head-on (HO) R-loops perturb normal fork progression.

(A) Replication products obtained on CD and HO templates with and without ClaI digestion post replication. (B) Conventional time course analyses of replication reactions on CD and HO templates. Time indicates minutes after addition of Mcm10 (origin firing).

Figure 2—figure supplement 2.. Characterization of replisome encounters with co-directional (CD) and head-on (HO) R-loops.

(A) Top: Schematic of products expected after digestion of full-length or restart replication products with Nt.BbvCI or Nb.BbvCI. Bottom: Denaturing gel analysis of replication products digested as indicated. (B) Replication products obtained on CD and HO templates in the absence or presence of Csm3-Tof1 and Mrc1 (CTM).

Figure 3.. dNTP pulse-chase analysis of fork progression through co-directional (CD) and head-on (HO) R-loops.

Pulse-chase time course analysis of replication reactions on CD (top) and HO (bottom) templates. Signal intensities of replication products were quantified and plotted as percentage of total signal.

Figure 4.. G-quadruplexes (G4s) and RNA:DNA hybrids pose impediment to leading strand synthesis that can be resolved by Pif1 or RNase H1, respectively.

(A) Purified proteins analyzed by SDS-PAGE and Coomassie stain. (B) Replication products obtained on co-directional (CD) or head-on (HO) templates in the absence or presence of RNase H1. (C) Replication products obtained on CD or HO templates in the absence or presence of RNase H1 and Pif1. (D) Replication products obtained on HO templates in the absence or presence of RNase H1 and pyridostatin. (E) Replication products obtained on CD templates in the absence or presence of RNase H1 and pyridostatin. ΔG4^810-828: Airn sequence containing deletion of G4 sequence at position 810–828.

Figure 4—figure supplement 1.. RNase H1 promotes fork passage specifically at co-directional (CD) R-loops.

(A) Schematic of replication products observed after replisome encounters with CD R-loops lacking G-quadruplexes (G4s) in the leading strand template. Treatment with RNase H1 removes the RNA:DNA hybrid, allowing completion of DNA replication. (B) Schematic of replication products observed after replisome bypass of CD R-loops harboring G4s in the leading strand template. Bypass of CD R-loops or fork passage in the presence of RNase H1 results in the uncoupling of leading strand synthesis at G4s, which promotes restart of leading strand synthesis by repriming.

Figure 4—figure supplement 2.. RNase H1 promotes fork passage specifically at co-directional (CD) R-loops.

(A) Schematic of replication products observed after replisome encounters with head-on (HO) R-loops harboring G-quadruplexes (G4s) in the displaced strand that block CMG progression. Stalling persists upon RNase H1 treatment. (B) Schematic of replication products observed after replisome encounters with HO R-loops harboring G4s in the displaced strand that are stabilized in single-stranded DNA, that is, when an RNA:DNA hybrid is present on the template strand, and that block leading strand polymerase but not CMG progression.

Figure 5.. CMG can unwind or translocate on RNA:DNA hybrids, while G-quadruplexes (G4s) can block DNA unwinding by CMG.

(A) Purified CMG. (B) Helicase assays with 40 bp forked DNA duplex preceded by 40 bp DNA (i + iii) or RNA:DNA (ii + iv) duplex. ★ indicates position of 5’-³² P label. Products were analyzed by native PAGE and autoradiography. (C) CMG helicase activity on 60 bp forked DNA (left) or RNA:DNA duplex (right). Plot shows average of two replicates. (D) CMG helicase activity on 60 bp substrate harboring wildtype (left) or mutant (right) G4 sequence on the template strand (‘lead’). (E) As (D), with wildtype (left) or mutant (right) G4 sequence on the non-template strand (‘lag’).

Figure 6.. G-quadruplexes (G4s) at co-directional (CD) R-loops can induce lagging strand gaps that can be resolved by Pif1.

(A) Replication products obtained on CD template in the absence or presence of Fen1/Cdc9 and RNase H1. (B) Schematic illustrating replication products observed in (A). (C) Replication products obtained on CD template in the presence of Fen1/Cdc9. RNase H1 was included and products were digested with Nt.BbvCI as indicated. (D) Replication products obtained on CD template in the presence of Fen1/Cdc9. Pif1 and RNase H1 were included as indicated.

Figure 7.. Both G-quadruplexes (G4s) and RNA:DNA hybrids cause lagging strand gaps at head-on (HO) R-loops.

(A) Replication products obtained on HO template in the absence or presence of Fen1/Cdc9 and RNase H1. (B) Replication products obtained on HO template harboring ΔG4^810-828 deletion in the absence or presence of Fen1/Cdc9 and RNase H1. (C) HO template lacking R-loop was replicated with Fen1/Cdc9 in the absence or presence of Pif1 or Pif1-E303Q. Signal intensities of Lag^US and Lag^DS were quantified and normalized to reactions without Pif1. (D) HO template replicated with Fen1/Cdc9 in the absence or presence of Pif1. (E) Schematic illustrating replication products observed in A-D.

Figure 7—figure supplement 1.. Both G-quadruplexes (G4s) and RNA:DNA hybrids cause lagging strand gaps at head-on (HO) R-loops.

(A) Denaturing gel analysis of replication products obtained in the presence of Fen1/Cdc9 on HO templates lacking R-loops were digested with ClaI (lanes 1 + 2), ClaI and MscI (lanes 3 + 4), or ClaI and BseRI (lanes 5 + 6). Lag^US and Lag^DS are sensitive to MscI and BseRI, respectively. (B) Schematic of expected products generated by digestion of full-length replication products. (C) Schematic of expected products generated by digestion of replication products containing a lagging strand gap.

Figure 8.. R-loop transcripts can prime leading strand restart after co-directional (CD) R-loop-replisome collisions.

(A) Replication products obtained on mock- or T4 polynucleotide kinase (PNK)-treated CD R-loop-containing templates in absence or presence RNase H1. (B) Products obtained on CD T4 PNK-treated R-loop templates in absence or presence of DDK. Relative signal intensity for restart product is quantified on the right. (C) Replication products obtained on mock- or T4 PNK-treated CD R-loop-containing templates in absence or presence RNase H1. (D) Schematic illustrating replication products observed in A. (E) RNase H1 titration into reactions with CD R-loop template. (F) Model for leading strand restart at R-loop transcript after replisome encounter with CD R-loop harboring 5’ RNA flap and RNA nick.

Figure 8—figure supplement 1.. R-loop transcripts can prime leading strand restart after co-directional (CD) R-loop-replisome collisions.

(A) Polymerase assay with T4 polynucleotide kinase (PNK)-treated CD R-loop template in presence of Pol α, Pol ε, or Pol δ, as indicated. All reactions include RFC, PCNA, RPA, dNTPs, NTPs, and α-[³²P]-dATP. The concentrations of reaction components are equivalent to those used in the replication assay. (B) Replication products obtained on RNase H1-treated CD R-loop templates with or without Pol δ. Reactions were performed either in the presence of 1 nM (lanes 1 + 2) or 0.06 nM (lanes 3 + 4) RNase H1 to induce resolution or nicking of the RNA:DNA hybrid, respectively. (C) Left: Purified RNase H2. Right: Replication products obtained on CD and head-on (HO) templates in the presence of sub-saturating levels of RNase H2.