RecD2 helicase limits replication fork stress in Bacillus subtilis

Abstract

DNA helicases have important roles in genome maintenance. The RecD helicase has been well studied as a component of the heterotrimeric RecBCD helicase-nuclease enzyme important for double-strand break repair in Escherichia coli. Interestingly, many bacteria lack RecBC and instead contain a RecD2 helicase, which is not known to function as part of a larger complex. Depending on the organism studied, RecD2 has been shown to provide resistance to a broad range of DNA-damaging agents while also contributing to mismatch repair (MMR). Here we investigated the importance of Bacillus subtilis RecD2 helicase to genome integrity. We show that deletion of recD2 confers a modest increase in the spontaneous mutation rate and that the mutational signature in ΔrecD2 cells is not consistent with an MMR defect, indicating a new function for RecD2 in B. subtilis. To further characterize the role of RecD2, we tested the deletion strain for sensitivity to DNA-damaging agents. We found that loss of RecD2 in B. subtilis sensitized cells to several DNA-damaging agents that can block or impair replication fork movement. Measurement of replication fork progression in vivo showed that forks collapse more frequently in ΔrecD2 cells, supporting the hypothesis that RecD2 is important for normal replication fork progression. Biochemical characterization of B. subtilis RecD2 showed that it is a 5'-3' helicase and that it directly binds single-stranded DNA binding protein. Together, our results highlight novel roles for RecD2 in DNA replication which help to maintain replication fork integrity during normal growth and when forks encounter DNA damage.

FIG 1

Functional domain alignment of RecD, RecD2, and UvrD. Sequences were aligned using the Clustal Omega program (http://www.ebi.ac.uk/Tools/msa/clustalo/), as described previously (65). Shown is a schematic representation of the indicated helicase domains. Conserved helicase domains and functions were assigned accordingly (16, 20, 24, 66). B. subtilis and B. anthracis RecD2 share 57% amino acid identity and 74% homology, whereas B. subtilis and D. radiodurans RecD2 share 28% amino acid identity and 50% homology. AA, amino acids.

FIG 2

Deletion of recD2 increases spontaneous mutagenesis in B. subtilis. Shown are bar graphs representing the spontaneous mutation rate (±95% confidence intervals) using the MSS maximum-likelihood method, as described previously (31, 63, 67–69). (A) Number of mutations per generation for wild-type strain PY79 and the ΔrecD2 and ΔmutSL strains from at least 50 independent cultures plated on rifampin; (B) number of mutations per generation of the PY79, ΔrecD2, and ΔmutSL strains from at least 20 independent cultures plated on trimethoprim; (C) number of mutations per generation of the indicated strains plated on rifampin. The data for PY79, ΔrecD2, and ΔmutSL shown in panel A are also shown in panel C.

FIG 3

Trimethoprim mutation spectra of B. subtilis wild-type, ΔrecD2, and ΔmutSL cells. The DNA sequence of the thyA gene of B. subtilis strain PY79 is shown. The spectra for the wild-type (green), ΔmutSL (purple), and ΔrecD2 (orange) strains are shown above the sequence. Filled triangles, insertion events; open triangles, deletions; solid line, duplications. The data shown here are also presented in more detail in Table S2 in the supplemental material.

FIG 4

The recD2 deletion confers sensitivity to mitomycin C, methyl methanesulfonate (MMC), phleomycin, and UV. (A) Serial dilutions of the indicated strains were plated on LB agar or LB agar with the indicated DNA-damaging agent. (B) Complementation of the ΔrecD2 strain with ectopic expression of recD2 from the amyE locus with the IPTG-inducible promoter (P_spac). Serial dilutions of the indicated strains were plated on LB agar, LB agar with phleomycin, or LB agar with phleomycin and 10 μM IPTG.

FIG 5

RecD2 stabilizes ongoing replication. (A) Coverage data from the right arm of the chromosome of an exponential-phase wild-type culture are plotted, with the red line denoting a linear fit to the data. (B) Replication initiation was halted in dnaB134 cells for 45 min, followed by sequencing of genomic DNA. The log₂ coverage of the right arm is plotted (black dots), and the red line shows a quadratic fit to the data. The data are the result of two independent experiments. (C) Same as in panel B, except that the genotype is ΔrecD2 dnaB134. The data are the result of two independent experiments. (D) The quadratic fits from panels B (solid line) and C (dashed line) are plotted together for comparison. (E) A table showing the fork collapse factor (FCF) for the right arm of the chromosome in each strain tested. CI, confidence interval.

FIG 6

RecD2 binds SSB and is a 5′-3′ helicase. (A) Immuno-dot blot of the RecD2 interaction with SSB. Each protein was serially diluted onto a nitrocellulose membrane and then incubated with either SSB (left) or RecD2 (right). The membrane was subsequently probed with polyclonal affinity-purified anti-SSB or anti-RecD2 antibodies, as described in Materials and Methods (30). (B) RecD2 unwinding of 5′ or 3′ tail-containing DNA substrates. RecD2 (0 to 2 nM) or RecD2(K373A) (2 nM) was incubated with the indicated DNA substrate for 25 min. (C) Quantification of substrate unwinding by RecD2. Percent unwinding by RecD2 was determined by dividing the intensity of the single-strand product band by the total intensity of the lane.