Figure 6. PcrA unwinds DNA-RNA hybrids in vitro and supresses R-loops in vivo (A) DNA and RNA substrates used for helicase assays.

Thick lines represent DNA strands and thin lines, RNA strands. The oligonucleotides used to form these substrates are shown in Table 7. (B) Quantification of unwinding as a function of PcrA concentration (in nM) for the 3′-tailed substrates shown in panel A. The substrate is only efficiently unwound if the longer of the two nucleic acids strands is DNA. Error bars show the standard deviation of at least three independent experiments. (C) The ATPase activity of PcrA is strongly stimulated by single-stranded DNA but not single-stranded RNA. Error bars show the standard deviation of at least three independent experiments. (D) Anti R-loop antibody (S9.6) dot blot for nucleic acid samples purified from three strains of B. subtilis. These strains contain an integrated expression cassette for either wild-type PcrA or a dominant negative form of PcrA (E224Q). The control strain (EV) contains an integrated but empty expression cassette. The S9.6 signal is normalised using methylene blue as a stain for all DNA. Note the high S9.6 signal for the strain expressing PcrA E224Q. (E) Quantification of four independent repeats of the experiment shown in (c). Error bars show the SEM. Expression of a dominant negative form of PcrA increases R-loop content (relative to DNA) in B. subtilis by ~2.5 fold. (F) Quantification of pulldown experiments of RNAP from B. subtilis cell extracts using biotinylated PcrA as bait and supplemented with purified PcrA WT or CTD. Addition of the CTD competes with WT PcrA to bind to RNAP. Error bars show the SEM of three independent repeats. (G) Relative R-loop levels in strains of B. subtilis expressing free CTD, a CTD mutant that interacts weakly with RNAP, or with a control expression cassette. A ΔrnhC strain is shown as a control for elevated R-loop levels. Error bars show the SEM of at least three independent experiments. In all panels, the statistical significance was determined using two-tailed Student’s t test (*p value < 0.05, **p value < 0.01, ***p value < 0.001, ****p value < 0.0001).

Figure 6—figure supplement 1. PcrA requires a 3′-ssDNA tail to unwind DNA duplexes.

(A) Representative helicase assays using 3′-tailed or duplex substrates consisting of annealed DNA (black) and RNA (red) oligonucleotides. The sequences of the oligonucleotides are in (Table 7). (B) Quantification of unwinding as a function of PcrA concentration (in nM) for 3′-tailed and fully duplex blunt-ended substrates (indicated by the prefix b). For the 3′-tailed substrates, the substrate is only efficiently unwound if the longer of the two nucleic acids strands is DNA. The fully duplex substrates are not unwound efficiently. Part of this data is reproduced from Figure 6 for comparison. Error bars show the standard deviation of at least three independent experiments.

Figure 6—figure supplement 2. Overexpression of PcrA E224Q causes growth defects in WT and Δmfd B. subtilis.

Growth curves for (A) WT strains or (B) Δmfd strains overexpressing wild-type PcrA or E224Q at 37°C. The overexpression was induced with 1 mM IPTG at the indicated times. The overexpression cassette was integrated in the codirectional orientation. Three biological replicates were averaged at each time point and error bars represent the standard error of the mean.

Figure 6—figure supplement 3. Overexpression of the PcrA CTD in B. subtilis and deletion of uvrD in E. coli increase R-loop levels in the cell.

(A) Representative anti-β subunit immunoblot (upper panel) and Coomassie-stained gel (lower panel) of a pulldown from B. subtilis extracts using biotinylated PcrA as bait. The gel shows that both free PcrA (full length) and free CTD can compete with the bait (tagged PcrA) to bind to RNAP. Where indicated, 1.5 µM untagged PcrA or CTD were added to the cell lysate. Quantification of these data is shown in Figure 6F. (B) and (C) Representative dot blots of B. subtilis CTD overexpression strains (as indicated) using the S9.6 antibody. Methylene blue was used as loading control. Quantification of these data is shown in Figure 6G. (D) and (E) Representative RNA/DNA hybrid dot blot of genomic DNA from the indicated E. coli strains and quantification. Error bars show the SEM of four independent experiments. In all panels, the statistical significance was determined using two-tailed Student’s t test (*p value < 0.05).