Skip to main content
NCBI home page
The EMBO Journal logoLink to The EMBO Journal
. 2007 Jul 19;26(16):3804–3814. doi: 10.1038/sj.emboj.7601804

UvrD controls the access of recombination proteins to blocked replication forks

Roxane Lestini 1, Bénédicte Michel 1,a
PMCID: PMC1952219  PMID: 17641684

Abstract

Blocked replication forks often need to be processed by recombination proteins prior to replication restart. In Escherichia coli, the UvrD repair helicase was recently shown to act at inactivated replication forks, where it counteracts a deleterious action of RecA. Using two mutants affected for different subunits of the polymerase III holoenzyme (Pol IIIh), we show here that the anti-RecA action of UvrD at blocked forks reflects two different activities of this enzyme. A defective UvrD mutant is able to antagonize RecA in cells affected for the Pol IIIh catalytic subunit DnaE. In this mutant, RecA action at blocked forks specifically requires the protein RarA (MgsA). We propose that UvrD prevents RecA binding, possibly by counteracting RarA. In contrast, at forks affected for the Pol IIIh clamp (DnaN), RarA is not required for RecA binding and the ATPase function of UvrD is essential to counteract RecA, supporting the idea that UvrD removes RecA from DNA. UvrD action on RecA is conserved in evolution as it can be performed in E. coli by the UvrD homologue from Bacillus subtilis, PcrA.

Keywords: DNA repair, Escherichia coli , RecA, RecQ, replication arrest

Introduction

During their progression, replication forks encounter obstacles that block their progression and eventually cause replisome disassembly. Replication impairment is now recognized as an important source of genetic instability in several organisms (Michel, 2000; Kolodner et al, 2002). Examples in mammalian systems include various disorders such as cancer (Shima et al, 2007). Escherichia coli chromosomal replication is catalyzed by a 10-subunit complex, the DNA polymerase III holoenzyme (Pol IIIh). Together with the replicative helicase (DnaB) and primase (DnaG), these complexes form the replisome (reviewed in Johnson and O'Donnell, 2005; O'Donnell, 2006). Replication restart requires PriA and other pre-primosomal proteins that reload the replicative helicase and, in turn, allow the re-assembly of a functional replication complex at inactivated replication forks (reviewed in Sandler and Marians, 2000). However, in several instances of replication arrest, replication does not restart directly and forks are first processed by recombination proteins (reviewed in Michel et al, 2004). Defects or errors in the processing of inactivated replication forks can have dramatic consequences as it can generate chromosome rearrangements. Our work aims at understanding the various pathways that process inactivated replication forks prior to replication restart. For this purpose, we used here mutants affected for two different subunits of Pol IIIh, the polymerase catalytic subunit DnaE and the clamp DnaN.

Any mutation or chemical that perturbs replication is likely to cause the formation of two types of abnormal structures: (i) single-strand gaps are formed when the lagging strand polymerase is impeded and are left behind the progressing replication fork and (ii) inactivated replication forks form when the leading strand polymerase is impeded (Langston and O'Donnell, 2006). These two types of structures co-exist in a cell population. Both contain single-strandeded DNA (ssDNA) and it is generally difficult to ascribe a deleterious phenotype caused by replication impairment to a defect in gap repair versus fork repair. In E. coli replication mutants, we can unambiguously determine when a protein acts at forks rather than gaps by studying a specific reaction, called replication fork reversal (RFR; Seigneur et al, 1998; Figure 1). The first step of the RFR reaction is the annealing of the leading- and lagging-strand ends at a replication fork to form a Holliday junction (HJ) with a DNA double-stranded end (dsDNA end). Recombination proteins reset the reversed fork. The dsDNA end is acted upon by RecBCD, a recombination complex specific for the repair of dsDNA breaks. RecBCD unwinds and degrades dsDNA up to a specific site named Chi, at which it shifts to a recombinase and loads the recombination enzyme RecA (Singleton et al, 2004, and references therein). We showed that RecBCD processes reversed forks either by degradation of the dsDNA end or by integration of this dsDNA end into the chromosome by homologous recombination, converting them into a structure from which replication restarts (Seigneur et al, 1998; pathways C and D in Figure 1). HJs made by homologous recombination or by fork reversal are branch-migrated by the RuvAB complex and resolved by the E. coli resolvase, the RuvABC complex (reviewed in West, 1997). Replication restarts from recombination intermediates or fork structures by the assembly of a new replisome via the action of PriA and its partner proteins (Figure 1). When RecBC is inactivated, the dsDNA end at reversed forks is not acted upon and resolution by RuvABC of the HJ made by fork reversal causes fork breakage (Figure 1E). Because resolution of HJs made by homologous recombination produces intact circular chromosomes, RuvABC only breaks chromosomes when it resolves an HJ made by fork reversal. Consequently, a modification of the level of RuvABC-dependent fork breakage by a mutation in a given gene indicates that the gene product affects RFR, hence acts at inactivated replication forks.

Figure 1.

Figure 1

Open in a new tab

Replication fork reversal model (left, pathways A–E) and model of action of UvrD (right, pathways F–G) (adapted from Seigneur et al, 1998 and Flores et al, 2005). In the first step (A), the replication fork is arrested by impairment of a replication protein, causing fork reversal. In the dnaEts mutant, this step is catalyzed by RuvAB. In both dnaEts and dnaNts mutants, it is also antagonized by RecQJFORA, which are, in turn, counteracted by UvrD. Consequently, the anti-fork reversal action of RecQJFORA can only be detected in uvrD mutants. The reversed fork forms an HJ with an open dsDNA end (two alternative representations of this structure are shown, open X and parallel stacked X). In Rec+ cells (pathways B–C), RecBCD initiates RecA-dependent homologous recombination at a Chi site present on the dsDNA end and the two HJs (one formed by reversal, one by homologous recombination) are resolved by RuvABC. Alternatively, if RecBCD encounters the HJ before a Chi site, or in the absence of RecA (B–D), the dsDNA end is degraded up to the HJ by the exonuclease V activity of RecBCD, restoring a fork structure. In both cases, replication restarts by a PriA-dependent process. In the absence of RecBC (E), resolution of the HJ by RuvABC causes fork breakage. In ΔuvrD mutants, RFR is prevented by the action of RecQJFORA; we proposed that RecA requires RecQ, RecJ and RecFOR to bind DNA at blocked forks (F, G), and that this binding prevents RFR but is counteracted by UvrD in UvrD+ cells. Continuous lines represent parental chromosome. Dashed lines represent newly synthesized strands. Gray circle represents RuvAB. Gray incised circle represents RecBCD. White incised circle represents RecJ. White crescent represents RecQ. Small white circles represent RecA.

The first step of the RFR reaction, conversion of a blocked fork into an HJ, is catalyzed by different pathways depending on the cause of replication arrest, suggesting that the processing of inactivated replication forks depends on proteins that remain or become associated with the fork after inactivation. Two RFR pathways have been characterized. Forks are reversed by a peculiar RecA-dependent reaction, which requires no known presynaptic protein in mutants affected for the replicative helicase DnaB (Seigneur et al, 2000). An RuvAB-dependent pathway acts in the rep helicase mutant and in two Pol IIIh mutants, dnaEts affected for the catalytic subunit of Pol IIIh and holD, affected for one of the subunits of the clamp loader (Baharoglu et al, 2006). A third, still unknown pathway, is active in the dnaNts mutant affected for the Pol IIIh clamp, in which RFR is RecA- and RuvAB-independent.

Monitoring RFR has recently allowed us to identify UvrD as a factor that participates in the resetting of inactivated replication forks (Flores et al, 2004, 2005). UvrD is a DNA helicase that belongs to the Super Family 1 of helicases (SF1; Gorbalenya and Koonin, 1993). It translocates on DNA in the 3′–5′ direction with a low processivity in vitro, and has been widely characterized for its role in nucleotide excision repair (NER) and mismatch repair (reviewed in Matson and Robertson, 2006). UvrD also acts during homologous recombination since uvrD-null mutants exhibit an increased frequency of homologous recombination (Zieg et al, 1978; Bierne et al, 1997), and a particular uvrD mutation (uvrD252, formerly recL152) inactivates homologous recombination in a recBC sbcBC background (Horii and Clark, 1973). Finally, UvrD is linked to replication: (i) it co-purifies with Pol III* (a Pol IIIh subcomplex that lacks the β-clamp DnaN) (Lahue et al, 1989), (ii) it is essential for the replication of rolling circle plasmids in E. coli (Bruand and Ehrlich, 2000), (iii) it is required for Tus removal from physiological replication arrest sites (Ter/Tus complexes) (Bidnenko et al, 2006) and (iv) it acts at inactivated replication forks in Pol IIIts mutants (Flores et al, 2005).

In the two Pol IIIh mutants dnaEts and dnaNts, RFR and viability at semi-permissive temperature require the presence of UvrD. However, UvrD is only needed when RecA and all the pre-synaptic proteins of the RecFOR recombination pathway are present: RecQ (a helicase that translocates in the 3′–5′ direction), RecJ (a 5′–3′ ssDNA exonuclease) and RecFOR, which promotes RecA binding to the ssDNA gaps. The need for UvrD in Pol IIIts mutants only when RecQ, RecJ, RecFOR and RecA are all present, led us to propose that UvrD antagonizes a deleterious concerted action of all these proteins: a RecQ- RecJ- RecFOR-dependent RecA binding to forks, which prevents RFR (Figure 1F and G). Reports that in vitro UvrD can undo RecA-ssDNA filaments and RecA-made recombination intermediates (Morel et al, 1993; Veaute et al, 2005) suggested that UvrD acts at forks after RecA binding, and allows RFR by clearing RecA from DNA at forks (Flores et al, 2005).

The action of UvrD during recombination and replication seems to be conserved in other species. The inactivation of SRS2, the yeast UvrD homologue, increases the frequency of homologous recombination and causes cell lethality in the presence of chemicals that perturb replication or in combination with mutations that affect replication (Fabre et al, 2002; Ooi et al, 2003; Pfander et al, 2005; Schmidt and Kolodner, 2006). Therefore, like UvrD in E. coli, SRS2 acts in Saccharomyces cerevisiae and in Schizosaccharomyces pombe to limit recombination and for survival after replication impairment. In Gram-positive bacteria, PcrA, the UvrD homologue, also acts in recombination and replication and PcrA from the distantly related bacteria Bacillus subtilis suppresses the NER and Tus blockage defects of a uvrD E. coli mutant (Petit et al, 1998; Petit and Ehrlich, 2002; Bidnenko et al, 2006). In support of the idea that there is a conservation of function of helicases in the UvrD family, we show here that the expression of PcrA also restores RFR in dnaEts ΔuvrD and dnaNts ΔuvrD cells.

Some of the mutations that inactivate the ATPase activity of uvrD in vitro impair growth of wild-type E. coli. One that did not affect wild-type E. coli growth is the uvrD-R284A mutation, but it could not be combined with dnaEts and dnaNts mutations (our unpublished results). In order to test whether the ATPase function of UvrD is required for its anti-RecA action at dnaEts and dnaNts-blocked forks, we used the uvrD252 mutant. The uvrD252 mutation is just adjacent to the Walker A motif (Gly30Asp), and in vitro the ATPase activity of the UvrD252 protein is strongly reduced, as well as DNA unwinding (Washburn and Kushner, 1993). UvrD252 mutant cells are defective for NER, but are capable of mismatch repair. Presumably this is due to MutL strongly activating the unwinding activity of UvrD (Hall et al, 1998; Yamaguchi et al, 1998; Guarne et al, 2004). We show here that, although UvrD252 is unable to dismantle RecA-ssDNA filaments or recombination intermediates in vivo, RFR occurs in a dnaEts uvrD252 mutant, but not in a dnaNts uvrD252 mutant. These results reveal the existence of two modes of action for UvrD at blocked forks, one that does not require the RecA removal activity of UvrD, and one that does, depending on the Pol IIIh subunit that is affected.

RarA (also called MgsA) has been proposed to act at replication forks (Barre et al, 2001). This protein is highly conserved in bacteria and eukaryotes. In yeast, Mgs1 plays a yet undetermined role in cell survival after replication impairment (Hishida et al, 2006; Vijeh Motlagh et al, 2006, and reference therein). In E. coli, RarA/MgsA is deleterious in the dnaEts mutant and its inactivation improves growth at semi-permissive temperature (Shibata et al, 2005). We show here that the inactivation of rarA restores RFR in a dnaEts uvrD mutant, indicating that RarA is required for RecQ, RecJ, RecFOR and RecA (called thereafter RecQJFORA) binding and/or action at dnaEts-blocked forks. In contrast, rarA inactivation does not restore RFR in dnaNts ΔuvrD mutants, indicating that RarA/MgsA is not required for RecQJFOR-promoted RecA binding at dnaNts-blocked forks. Interestingly, the requirement for RarA correlates with a role of UvrD, independent of its ATPase activity.

Results

The helicase function of UvrD is not required for RFR in the dnaEts mutant

The occurrence of RFR upon replication impairment can be detected by measuring chromosome breakage, as resolution by RuvABC of the HJs formed at reversed forks leads to fork breakage in a recBC mutant context (Seigneur et al, 1998). In order to quantify fork breakage in vivo, we measure the proportion of broken chromosomes in a cell population by pulse field gel electrophoresis (PFGE). dnaNts- and dnaEts-dependent fork-breakage is maximum at 37 and 42°C, respectively, and is decreased to the level of replication proficient cells by uvrD inactivation (Figure 2; Flores et al, 2004, 2005; Supplementary Tables S2 and S3; Supplementary Figure S1). Therefore, UvrD is required for RFR to occur in these mutants.

Figure 2.

Figure 2

Open in a new tab

UvrD252 allows RFR in the presence of RecQJFORA at dnaEts-blocked forks, but not at dnaNts-blocked forks. The histograms indicate the percentage of linear DNA measured by PFGE in cultures propagated for three hours at restrictive temperature. Scale bars indicate the standard deviation. (A) dnaEts mutants at 42°C. From left to right: dnaEts recBCts (JJC1983), dnaEts recBCts ΔuvrD (JJC2007), dnaEts recBCts uvrD252 (JJC2545 and JJC2566), dnaEts uvrD252 ΔrecA [pBR-Gam] (JJC2987 [pBR-Gam], dnaEts recBCts uvrD252 ΔruvABC rusA (JJC3911 and JJC3922, the rusA gene is inactive in wild-type E. coli and rusA designates here a mutation that activates the gene and consequently rusA mutant strains express the RusA resolvase), dnaEts recBCts uvrD252 mutL (JJC2567 and JJC3922), dnaEts recBCts uvrD252 mutS (JJC3141). (B) dnaNts mutants at 37°C. From left to right: dnaNts recBCts (JJC1982 and JJC2057), dnaNts recBCts ΔuvrD (JJC2006), dnaNts recBCts uvrD252 (JJC2554 and JJC2656), JJC2554 or JJC2556 carrying pGB-UvrD+, JJC2554 or JJC2556 carrying pGB-RuvABC+, dnaNts uvrD252 ΔrecA [pBR-Gam] (JJC3038 [pBR-Gam]), dnaNts recBCts uvrD252 recO (JJC2577), dnaNts recBCts uvrD252 recO ΔruvABC (JJC3809). Breakage in dnaEts recBCts (JJC1983), dnaEts recBCts ΔuvrD (JJC2007), dnaNts recBCts (JJC1982 and JJC2057), dnaNts recBCts ΔuvrD (JJC2006) were previously published (Grompone et al, 2002; Flores et al, 2004), and were reproduced here in parallel with the uvrD252 strains.

We used the uvrD252 allele to test whether the helicase activity of UvrD is required for RFR at Pol IIIts-blocked forks. The level of fork breakage in dnaEts recBCts uvrD252 cells was not significantly different from that in dnaEts recBCts UvrD+ cells (P=0.4), and significantly higher than that in the dnaEts recBCts ΔuvrD mutant (P<0.001; Figure 2A; Supplementary Table S2; Supplementary Figure S1). As a control, the ΔuvrD allele was introduced by P1 transduction into in the dnaEts recBCts uvrD252 mutant; fork breakage decreased in this ΔuvrD construct, confirming that the presence of the UvrD252 protein allows fork breakage (JJC2545 ΔuvrD; Supplementary Table S2). Using cells that express the RusA resolvase but lack RuvABC, we previously showed that, in a dnaEts mutant, RFR requires the HJ branch migration activity of the RuvAB complex (Baharoglu et al, 2006). The inactivation of DnaE did not cause fork breakage when RusA was expressed in the dnaEts uvrD252 recBCts ΔruvABC mutant, indicating that RFR requires the branch migration activity of RuvAB in dnaEts uvrD252 mutant as in dnaEts UvrD+ cells (Figure 2A; Supplementary Table S2). RFR is independent of RecA in the dnaEts mutant and, as expected, recA inactivation did not modify the high level of fork breakage in the dnaEts recBCts uvrD252 mutant (Figure 2A; Supplementary Table S2). In conclusion, dnaEts recBCts cells behave similarly when they are UvrD+ or combined with the uvrD252 allele, namely they undergo a high level of fork breakage, which requires the branch migration activity of RuvAB. The high level of fork breakage in uvrD252 cells that express all RecQJFORA proteins, indicates that, as UvrD wild-type protein, UvrD252 is able to allow RuvAB-dependent RFR in the presence of RecQJFORA. In other words, the uvrD252 mutation does not inactivate the function of UvrD that allows it to counteract the deleterious action of RecQJFORA at dnaEts-blocked forks.

UvrD252 is active in mismatch repair presumably because it is activated by specific interactions with MutL; therefore we tested whether the activity of this mutant protein at dnaEts-blocked forks requires MutL. The inactivation of mutL or mutS did not modify the level of fork breakage in the dnaEts recBCts uvrD252 mutant (P=0.43; Figure 2A; Supplementary Table S2), indicating that MutL or MutS are not required for the action of UvrD252 at dnaEts-blocked forks.

The helicase function of UvrD is required for RFR in the dnaNts mutant

The uvrD252 mutation was used to determine whether the ATPase function of UvrD is required for RFR at dnaNts-blocked forks. The level of fork breakage at 37°C was similar in dnaNts recBCts uvrD252 to that of the dnaNts recBCts ΔuvrD (P=0.2), however, significantly lower than in dnaNts recBCts UvrD+ cells (P<0.001; Figure 2B; Supplementary Table S3; Supplementary Figure S1). As expected, fork breakage was restored in dnaNts recBCts uvrD252 mutant by the introduction of the wild-type uvrD gene in a plasmid. Moreover, the introduction of additional copies of RuvABC did not modify the level of breakage, indicating that the low level of fork breakage does not result from a limiting amount of the resolvase RuvABC in this strain (Figure 2B; Supplementary Table S3). A high level of fork breakage was restored by the inactivation of recQ, recJ, recO or recA (Figure 2B; Supplementary Table S3). We ensured that chromosome breakage in dnaNts recBCts uvrD252 recO cells results from RFR by testing that it required the resolvase activity of RuvABC (Figure 2B; Supplementary Table S3). These results indicate that, as previously observed in dnaNts ΔuvrD cells, RFR is prevented in the dnaNts uvrD252 mutant by the presence of RecQJFORA. Therefore, the function of UvrD that allows RFR at dnaNts-blocked forks in the presence of RecQJFORA is inactivated by the uvrD252 mutation, suggesting a requirement for the helicase or the translocase function of UvrD to counteract RecQJFORA in this replication mutant.

The RFR defect of the uvrD mutant is suppressed by Bacillus subtilis PcrA

UvrD is largely conserved in bacteria and, in order to test whether its capacity to counteract RecA at blocked forks is conserved, we tested whether the expression of the PcrA protein from B. subtilis can complement the RFR defect of dnaEts ΔuvrD and dnaNts ΔuvrD mutants. A dnaEts recBCts ΔuvrD PcrA+ mutant was constructed. In this strain, the pcrA gene is inserted as a single copy in the chromosome and, although it is fused to the strong Pspac promoter, it is partially repressed by the adjacent LacI repressor. Experiments were performed in the absence of the IPTG inducer, so that PcrA is unlikely to be over-expressed, but complete induction of pSpac by IPTG did not affect results (Supplementary Tables S2 and S3, and data not shown).

Analysis of chromosomes by PFGE showed that the expression of PcrA restored fork breakage, hence fully suppressed the ΔuvrD defect (Figure 3; Supplementary Table S2). Fork breakage in dnaEts recBCts ΔuvrD PcrA+ mutant was unaffected by the inactivation of recO or recA and required the HJ branch migration activity of RuvAB (no dnaEts-dependent fork breakage was observed in ruvABC rusA1 mutants, in which the resolution of HJ is performed by the RusA resolvase; Figure 3; Supplementary Table S2). These results mimic those observed in dnaEts recBCts UvrD+ cells (Flores et al, 2005; Baharoglu et al, 2006), indicating that at dnaEts-blocked forks PcrA is able to catalyze an anti-RecQJFORA action, thereby allowing RuvAB-dependent RFR.

Figure 3.

Figure 3

Open in a new tab

Expression of PcrA suppresses the RFR defect of dnaEts ΔuvrD and dnaNts ΔuvrD cells. (A) dnaEts mutants at 42°C. From left to right: dnaEts recBCts ΔuvrD PcrA+ (JJC2692, fork breakage is not significantly different from dnaEts recBCts cells, P=0.9), dnaEts recBCts ΔuvrD PcrA+ ruvAB RusA+ (JJC3156), dnaEts recBCts ΔuvrD PcrA+ recO (JJC2719), dnaEts ΔuvrD PcrA+ recA [pBR-Gam] (JJC3351 [pBR-Gam]). (B) dnaNts mutants at 37°C. From left to right: dnaNts recBCts ΔuvrD PcrA+ (JJC3326 and JJC3502, fork breakage is not significantly different from dnaNts recBCts cells, P=0.8), dnaNts recBCts ΔuvrD PcrA+ ruvAB (JJC3724), dnaNts recBCts ΔuvrD PcrA+ recO (JJC2720), dnaNts recBCts ΔuvrD PcrA+ recJ (JJC3509), dnaNts recBCts ΔuvrD PcrA+ recQ (JJC3339), dnaNts recBCts ΔuvrD PcrA+ ΔrecA (JJC3708).

The expression of PcrA also fully suppressed the ΔuvrD defect in dnaNts recBCts ΔuvrD PcrA+ cells by restoring fork breakage (Figure 3; Supplementary Table S3). As expected, the inactivation of the resolvase RuvABC drastically reduced fork breakage, indicating that in a ΔuvrD PcrA+, as in a UvrD+ context, chromosome breakage upon dnaN inactivation is catalyzed by HJ resolution, hence results from RFR. RFR was independent of RecQ, RecJ, RecO and RecA in dnaNts uvrD PcrA+, as in dnaNts UvrD+ cells, suggesting that complementation of the uvrD defect by PcrA may not modify the mode of RFR (Figure 3; Supplementary Table S3). In addition to restoring RFR in both dnaEts ΔuvrD and dnaNts ΔuvrD mutants, PcrA also restored the viability of Pol IIIts ΔuvrD cells at 37°C to the level of the Pol IIIts single mutants (data not shown).

UvrD252 does not dismantle recombination intermediates

UvrD is capable in vitro of two anti-RecA actions: unwinding of recombination intermediates to restore substrate molecules and dismantling of RecA filaments (Morel et al, 1993; Veaute et al, 2005). In vivo, the hyper-recombination phenotype of the ΔuvrD mutant presumably results from the loss of these activities (Zieg et al, 1978). We previously reported that ΔuvrD ruvAB cells are lethal, and this lethality is suppressed by the inactivation of recFOR or recA genes (Flores et al, 2005). Since RecFOR promotes RecA binding to gaps (Kuzminov, 1999; Morimatsu and Kowalczykowski, 2003), we proposed that the lethality of ΔuvrD ruvAB cells results from an increased formation of ssDNA gaps in the ΔuvrD mutant, which are repaired by RecFOR-dependent homologous recombination, increasing the need for HJ resolution by RuvABC. In the absence of RecFOR, HJ resolution is not needed because gaps are presumably repaired by gap filling. The ΔuvrD ruvAB recQ triple mutant could not be constructed, either by P1 transduction of ruvABC mutation in ΔuvrD recQ (Table I) or by P1 transduction of ΔuvrD in a recQ ruvAB mutant (JJC2776 Supplementary Table SI, data not shown). The lethality of the ΔuvrD ruvAB recQ triple mutant indicates that gap recombinational repair in ΔuvrD mutant is RecQ-independent. It should be noted that uvrD ruv (Bierne et al, 1997) and uvrD ruv recQ (Magner et al, 2007) mutants were observed to be viable when constructed in E. coli backgrounds other than the AB1157 context used in the present work.

Table 1.

RecFOR-dependent hyper-recombination in uvrD mutants make them incompatible with ruvAB inactivation

Receptor strain ΔruvABC∷CmR clones among TetR eda∷Tn10 P1 transductants
Strain Genotype Number of CmR/TetR transductants Ratio
JJC40 wild-type 47/83 0.57
JJC1628 recQ∷Tn3 47/69 0.68
JJC4174 ΔrecQ∷Cm 35/59 0.59
JJC2642 ΔuvrD 0/75 <0.01
JJC2821 ΔuvrD recO 32/51 0.63
JJC3850/JJC2603 ΔuvrD recQ∷Tn3 0/131 <0.01
JJC4152 ΔuvrD ΔrecQ∷Cm 0/40 <0.02
JJC2530 uvrD252 0/133 <0.01
JJC2983 uvrD252 recO 48/63 0.76
JJC3129 uvrD252 recQ∷Tn3 0/93 <0.02
JJC3686 uvrD252 ΔrecQ∷Cm 0/40 <0.02
JJC2673 ΔuvrD PcrA+ 31/52 0.60
Open in a new tab

eda and ruv genes are about 12 kb apart. ΔruvABC and Ruv+ transductants colonies were of similar size. ΔruvABC∷CmR clones were identified by picking eda∷Tn10 TetR transductants clones on chloramphenicol-containing medium, and checked by PCR using oligonucleotides that flank the ruvABC region and by UV irradiation. For JJC2673, JJC4174, JJC4152 and JJC3686, which are CmR, ΔruvABC∷CmR clones, were identified by PCR. With the four uvrD recQ double mutants, some micro-colonies appeared besides the Ruv+ TetR transductants. These micro-colonies might be abortive ΔruvABC co-transductants, but this could not be checked because these micro-colonies could not be streaked or cultured.

Like ΔuvrD, the uvrD252 mutation was co-lethal with ruvAB inactivation in our background. Viability was restored when gap recombinational repair was prevented by a recO mutation (Table I, and JJC3809 in Supplementary Table S3), when the RusA resolvase is activated (see JJC3911 in Supplementary Table S2), but not by recQ inactivation (Table I). The requirement for RuvABC in the uvrD252 mutant suggests an increased level of gap recombinational repair. Furthermore, conjugational recombination was largely increased in the uvrD252 mutant (Supplementary Table S4). Altogether, the observation that recombination is increased in uvrD252 as in ΔuvrD mutant indicates that the ATPase function of UvrD is required for the dismantling of recombination intermediates. In contrast, the expression of PcrA suppressed the lethality of ΔuvrD ruvAB double mutant (Table I). Therefore PcrA suppresses the spontaneous hyper-recombination phenotype of ΔuvrD cells.

UvrD252 does not remove RecA from SOS-inducing ssDNA gaps

We tested the capacity of UvrD252 and PcrA to undo ssDNA-RecA filaments in vivo by measuring SOS induction using an sfiA::lacZ fusion (Table II). SOS induction and an increased formation of RecA-GFP foci were previously reported in uvrD mutants (Ossanna and Mount, 1989; SaiSree et al, 2000; Centore and Sandler, 2007). The SOS response was slightly less induced in the uvrD252 mutant than in ΔuvrD cells, suggesting either the formation of less ssDNA substrates or a weak ability of UvrD252 to dismantle RecA filaments in a wild-type context (Table II). In both ΔuvrD and uvrD252 mutants, SOS induction was RecQ-independent and partly RecF-dependent, indicating that it results, in part, from RecQ-independent RecA binding to ssDNA gaps. The remaining SOS induction is presumably RecBC-dependent, but uvrD recB mutants are too poorly viable to be tested. SOS induction was decreased to the wild-type level by the expression of PcrA in ΔuvrD PcrA+ cells, indicating that the B. subtilis protein PcrA either dismantles ssDNA-RecA filaments in E. coli ΔuvrD, or complements the uvrD defects that originally cause their formation (Table II).

Table 2.

SOS induction due to Pol IIIts and uvrD mutations

Strain Relevant genotype 30°C 37°C
    Miller units Ratio Miller units Ratio
JJC2255 Wild-type 44±1 1 37.6±9.6 1
JJC2988 ΔuvrD 284±3 6.5 297±52 7.8
JJC2989 uvrD252 216±12 4.9 183±35 4.9
JJC3066 recO ND ND 81±12 2.2
JJC3101 ΔuvrD recF ND ND 132±22 3.5
JJC3102 uvrD252 recF ND ND 114±7 3
JJC3114 recQ ND ND 52±20 1.4
JJC3108 ΔuvrD recQ ND ND 258±9 6.8
JJC3912 uvrD252 recQ ND ND 141±37 3.7
JJC2985 ΔuvrD PcrA+ ND ND 39±6 0.9
JJC3033 rarA ND ND 83±6 2.2
           
JJC2991 dnaEts 70±12 1 427±113 1
JJC2993 dnaEts ΔuvrD 1500±630 21 1518±551 3.5
JJC2995 dnaEts uvrD252 1653±370 24 1705±507 4
  dnaEts uvrD252 [pGB2] 1405±344 20 1922±6.4 4.5
  dnaEts uvrD252 [pGB-UvrD+] 76±17 1.1 200±23 0.5
JJC3069 dnaEts recO 98±21 1.4 127±19 0.3
JJC3075 dnaEts ΔuvrD recO 99±27 1.4 121±22 0.28
JJC3081 dnaEts uvrD252 recO 177±32 2.5 129±77 0.3
JJC3079 dnaEts recQ 94±27 1.3 473±70 1
JJC3068 dnaEts ΔuvrD recQ 646±134 9.2 1270±241 3
JJC3919 dnaEts uvrD252 recQ 750±132 10.7 1353±175 3.2
  dnaEts uvrD252 recQ [pGB2] 460±4 6.6 1448±47 3.4
  dnaEts uvrD252 recQ [pGB-UvrD+] 44±1 0.6 169±8 0.4
JJC2997 dnaEts ΔuvrD PcrA+ 74±32 1 548±139 1.3
JJC3034 dnaEts rarA 97±18 1.4 331±16 0.7
           
JJC2992 dnaNts 79±11 1 522±146 1
JJC2994 dnaNts ΔuvrD 1642±287 21 2361±537 4.5
JJC2996 dnaNts uvrD252 994±124 12.6 2113±550 4.0
  dnaNts uvrD252 [pGB2] 1101±230 14 1742±338 3.3
  dnaNts uvrD252 [pGB-UvrD+] 99±13 1.2 305±52 0.6
JJC3077 dnaNts recO 150±7.6 2 209±21 0.4
JJC3076 dnaNts ΔuvrD recO 149±8 1.9 207±23 0.4
JJC3078 dnaNts uvrD252 recO 238±21 3 349±23 0.7
JJC3080 dnaNts recQ 92±12 1.2 610±68 1.1
JJC3084 dnaNts ΔuvrD recQ 828±170 10.5 1602±591 3.1
JJC3083 dnaNts uvrD252 recQ 1214±77 15 2451±339 4.7
  dnaNts uvrD252 recQ [pGB2] 1310±49 16 2442±380 4.7
  dnaNts uvrD252 recQ [pGB-UvrD+] 104±8 1.3 381±11 0.7
JJC2998 dnaNts ΔuvrD PcrA+ 95±16 1.2 813±45 1.5
JJC3035 dnaNts rarA 96±15 1.2 567±39 1.1
Open in a new tab

Since the uvrD252 mutation allows RFR in dnaEts but not in dnaNts cells (Figure 2), we tested whether SOS induction is differentially affected by the uvrD252 allele in these two replication mutants, by comparing SOS induction in dnaEts (or dnaNts) ΔuvrD and dnaEts (or dnaNts) uvrD252 mutants. Because they behaved similarly, the dnaEts and dnaNts mutants are referred to as Pol IIIts below.

As previously reported, Pol IIIts single mutants are induced for SOS at 37°C (Table II; Flores et al, 2005). This SOS induction is shown here to depend entirely on RecO and not on RecQ, and therefore to result from RecQ-independent RecA filaments formed at ssDNA gaps. Since our previous results indicate that RecQ is required for prevention of RFR by RecFOR and RecA (Flores et al, 2005), we infer that SOS-inducing gaps are not at blocked forks but elsewhere on the chromosome. Altogether these results points to the formation of two types of RecFOR-dependent RecA filaments in Pol IIIts cells: RecQ-dependent at forks, and RecQ-independent elsewhere on the chromosome.

SOS induction by Pol IIIts and either ΔuvrD or uvrD252 mutations was highly synergistic (Table II). SOS was induced even at the permissive temperature of 30°C in all Pol IIIts uvrD double mutants, indicating that (as often observed with ts mutants) replication is not fully functional even in conditions permissive for growth. SOS induction was entirely RecO-dependent, showing that it results from RecA binding to ssDNA gaps. However, it was decreased at most two-fold, and only at 30°C, by recQ inactivation. Since RecQ act together with RecFOR at forks as deduced from fork breakage measurements (Flores et al, 2005; Supplementary Table S3), the very weak effect of recQ inactivation compared to recO on SOS induction indicates that most of the SOS-inducing RecA filaments are not at forks, but are rather on gaps left behind progressing replication forks on daughter chromosomes. Importantly, the high level of SOS induction in the two Pol IIIts uvrD252 mutants clearly indicates that the UvrD252 mutant protein cannot remove RecA from ssDNA in Pol IIIts contexts. Since UvrD252 is nevertheless able to prevent RecA deleterious action at dnaEts-blocked forks, this finding suggests that the UvrD252 (and possibly UvrD) target is not ssDNA-bound RecA filaments.

PcrA decreased the level of SOS induction in dnaEts ΔuvrD and dnaNts ΔuvrD nearly to the level of dnaEts and dnaNts single mutants. This result indicates that in both dnaEts ΔuvrD and in dnaNts ΔuvrD mutants, PcrA is able to undo or prevent the formation of the RecA filaments caused by the ΔuvrD mutation, but not those resulting from the Pol III defect (Table II). This result is compatible with the idea that PcrA allows RFR in a ΔuvrD context by removing RecA from DNA.

MgsA/RarA is needed for RecQJFORA action at dnaEts-blocked fork

dnaEts cells grow slowly at 37°C and are killed at 38°C. These growth defects at these semi-permissive temperatures result from the action of RecQJFORA proteins, as they are suppressed by the inactivation of any of recQJFORA genes (Hishida et al, 2004; Flores et al, 2005). The growth defect caused by the presence of RecQJFORA is strongly amplified in the absence of UvrD, as Pol IIIts ΔuvrD mutants are lethal at 37°C and this lethality is suppressed by the inactivation of any of the recQJFORA genes (Flores et al, 2005). Like the inactivation of recQ, recJ, recFOR or recA, the inactivation of rarA (mgsA) allows growth of dnaEts single mutant at 38°C and improves its viability at 37°C (Hishida et al, 2006; Figure 4). Nevertheless, as with recQ inactivation, the rarA mutation did not suppress the SOS induction due to the dnaEts or dnaNts mutations (Table II). Furthermore, fork breakage was unaffected by rarA inactivation in dnaEts recBCts or in dnaNts recBCts cells and remained dependent on the branch migration activity of RuvAB in dnaEts recBCts (Figure 5; Supplementary Table S5), indicating that rarA does not affect RFR in these mutants, provided UvrD is active.

Figure 4.

Figure 4

Open in a new tab

rarA inactivation specifically improves the viability of dnaEts and dnaEts uvrD cells. Cells propagated at 30°C for 2 h were shifted to 37°C (time 0), and appropriate dilutions were plated at the indicated times; plates were counted after 48 h incubation at 30°C. (A) dnaEts rarA (JJC2910, gray squares), dnaEts (JJC1954, white squares), dnaEts uvrD rarA (JJC2918, gray circles), dnaEts uvrD recQ (JJC2748, triangles), dnaEts uvrD (JJC2697, crosses and dashed line). (B) dnaNts (JJC2434, white squares); dnaNts rarA (JJC2911, gray squares), dnaNts uvrD recQ (JJC2759, triangles), dnaNts uvrD rarA (JJC2919, gray circles), dnaNts uvrD (JJC2698, stars and dashed line). Growth curves of dnaEts, dnaEts uvrD, dnaEts uvrD recQ, dnaNts, dnaNts uvrD, dnaNts uvrD recQ were previously published (Flores et al, 2005), and were reproduced here in parallel with the rarA strains.

Figure 5.

Figure 5

Open in a new tab

rarA inactivation suppresses the RFR defect in dnaEts ΔuvrD but not dnaNts ΔuvrD cells. Dark bars, dnaEts recBCts rarA, gray bars, dnaNts recBCts rarA. From left to right: dnaEts recBCts rarA (JJC1744 and JJC2912), dnaEts recBCts rarA ruvAB (JJC1752), dnaEts recBCts rarA ruvAB RusA+ (JJC3055), dnaEts recBCts rarA uvrD (JJC2932), dnaNts recBCts rarA (JJC1743, JJC2913 and JJC3729), dnaNts recBCts rarA ruvAB (JJC3781), dnaNts recBCts rarA uvrD (JJC2920), dnaNts recBCts rarA uvrD recO (JJC3057). dnaEts recBCts rarA and dnaEts recBCts rarA ΔuvrD are not significantly different (P=0.6), whereas dnaNts recBCts rarA and dnaNts recBCts rarA ΔuvrD are (P<0.001).

To gain more insight in the action of RarA (MgsA), we tested the effects of the rarA mutation in Pol IIIts uvrD double mutants. The inactivation of mgsA/rarA improved the viability of dnaEts uvrD cells at 37°C to the same level as the inactivation of recQ (Figure 4A), but less than the inactivation of recFOR, which restored complete growth (Flores et al, 2005). Interestingly, rarA inactivation restored fork breakage in the dnaEts recBCts uvrD mutant (Figure 5; Supplementary Table S5). This result indicates that RarA (together with RecQJFORA) prevents RFR at dnaEts-blocked forks in the absence of UvrD, suggesting that RarA is required for RecQJFOR action at dnaEts-blocked forks.

The action of RarA in the dnaNts uvrD mutant was studied at 37°C by measuring viability of dnaNts rarA uvrD, and measuring fork breakage in dnaNts rarA recBCts uvrD cells. Interestingly, in contrast with results in the dnaEts context, rarA inactivation did not improve growth of dnaNts ΔuvrD cells (Figure 4B), and did not restore fork breakage in a dnaNts recBCts uvrD mutant (Figure 5). Therefore, the action of rarA is specific for dnaEts-blocked forks.

Discussion

Knowledge of the reactions that occur upon replication impairment is crucial for a complete understanding of the links between replication and genome stability. The natural causes of spontaneous replication arrests are unknown and are likely to be multiple, as DNA and replication proteins are susceptible to be affected. In the present work, we analyzed the action of UvrD and RarA (MgsA) at replication forks inactivated by a Pol IIIts mutation. When Pol IIIts mutants are propagated at the semi-permissive temperature of 37°C they undergo stochastic replication fork arrest (Grompone et al, 2002; Flores et al, 2004, 2005). In addition, this defective replication also forms of ssDNA gaps that can be distinguished from blocked forks because RecA binds them in a RecQ-independent manner (Table II). Presumably, depending on whether the lagging or the leading strand polymerase is inactivated first, a gap on the lagging strand template or a blocked fork will form, and both are present in a cell population. uvrD mutations also cause the formation of ssDNA gaps, which are revealed by SOS induction and recombination increase. Regardless of whether a uvrD-null or a uvrD252 mutation is used, the SOS response is highly induced in Pol IIIts uvrD double mutants. Therefore, in agreement with its strong ATPase defect in vitro, the UvrD252 protein cannot remove RecA from ssDNA in vivo. The use of this mutant allows us to demonstrate the existence of two different modes of action of the helicase UvrD that antagonize two different pathways for RecQJFORA access to blocked forks. When replication forks are arrested by the inactivation of the catalytic subunit of Pol IIIh DnaE, RarA participates in the blockage of RFR in the absence of UvrD (together with RecQJFORA; Flores et al, 2005), and the ATPase-deficient UvrD mutant is able to overcome this blockage. In cells deficient for the β-clamp DnaN, RecQJFORA does not require RarA for blocking RFR, and the ATPase activity of UvrD is required to counteract their deleterious action.

RarA (MgsA) is required for RecQ action at certain blocked forks

rarA is coexpressed in operon with ftsK, which encodes a multifunctional septum protein required for chromosome segregation and chromosome dimer resolution. It shares homology with the dnaX-encoded subunits of Pol IIIh and with RuvB (Barre et al, 2001). RarA is a highly conserved protein; the S. cerevisiae and E. coli proteins share 40% identity (Barre et al, 2001). This high level of conservation suggests an important role for the protein. Using GFP-RarA and GFP-RecQ fusion proteins, it was observed that RarA and RecQ are both located in the region of the replication factory in growing cells (Sherratt et al, 2004). Nevertheless, the function of RarA in E. coli and of its counterpart Mgs1 in yeast has remained elusive. Both the rarA mutation in E. coli and the Mgs1 mutation in S. cerevisiae improve the growth of mutants affected for the replicative polymerase (Hishida et al, 2002; Vijeh Motlagh et al, 2006). Mgs1 protein interacts with the Pol 31, one of the subunits of the yeast replicative polymerase and with the PCNA sliding clamp (the DnaN eukaryote homologue) (Hishida et al, 2006; Vijeh Motlagh et al, 2006). Mgs1 from human cells (termed WHIP) interacts with the RecQ human homologue, the Werner helicase (Kawabe et al, 2001). The similarity of phenotypes conferred by rarA or recQJFORA inactivation in the dnaEts mutant suggests that RarA is required for RecQJFOR-dependent RecA binding to blocked forks. RarA might interact with RecQ and load it onto replication fork. However, this hypothesis is difficult to reconcile with the observation that RarA is not required for RecQ loading at dnaNts-blocked forks and for RecQ-dependent homologous recombination in recBC sbcBC cells (Nakayama et al, 1985, our unpublished results). We instead favor a model in which RarA binds to dnaEts-blocked forks and, in the absence of UvrD, renders forks accessible to RecQ. Clearly, the structure of DNA and/or the proteins present are different after replication arrest in dnaEts and dnaNts mutants. We can speculate that RarA is required for RecQ binding only in the dnaEts uvrD mutant, because the disassembly of Pol IIIh is partial in this mutant, preventing RecQ access while still allowing RuvAB to revert forks. In this scenario, DnaN could be among the Pol IIIh subunits that remain bound to DNA in a dnaEts mutant, and its absence from dnaNts-blocked forks would render RarA dispensable for RecQ binding.

The helicase activity of UvrD is not required when RarA and RecQ act in concert

The existence of two pathways, RarA-dependent and RarA-independent, for RecQJFORA binding correlates with two different modes of action of UvrD. The existence of these two modes of action are revealed by the use of the uvrD252 allele, which affects RFR differently in the two Pol IIIts mutants tested. The weak residual ATPase activity of the UvrD252 enzyme detected in vitro is only significant in vivo when the helicase activity of UvrD is activated by MutL during mismatch repair. We observe here that the uvrD252 mutant is completely defective for three anti-RecA actions of UvrD, which are its capacity to limit homologous recombination and SOS induction and its capacity to antagonize RecA-dependent blockage of RFR in the dnaNts mutant. Nevertheless, UvrD252 fully allows RFR in the presence of all RecQJFORA proteins in a dnaEts mutant. It is possible that the UvrD252 ATPase function is specifically activated by an unknown protein other than MutLS, after RecA binding to inactivated forks in a dnaEts mutant. Although we cannot formally exclude this possibility, it implies the presence of different factors after RecQJFOR-dependent RecA binding to inactivated replication forks. Instead, we favor an alternative model (Figure 6), which suggests that UvrD does not need its helicase activity to allow RFR at dnaEts-blocked fork, because it acts before RecA binding. The correlation between UvrD252 and RarA action leads us to speculate that UvrD (or UvrD252) could prevent RecA binding by antagonizing RarA. In contrast, in the dnaNts mutant, the structure of the forks and the associated proteins are such that RecQ gains access to forks and promotes RecA binding regardless of the presence of UvrD and RarA. Nevertheless, UvrD can then dislodge RecA from blocked forks DNA, provided its ATPase is functional (Figure 6).

Figure 6.

Figure 6

Open in a new tab

Model of action of UvrD in dnaEts and dnaNts. In the dnaEts mutant (left part of the figure) inactivated replication forks can be channeled to two different pathways depending on the presence (but not on the activity) of UvrD. UvrD wild-type or ATPase-deficient proteins prevent RarA and the presynaptic proteins RecQJFOR from promoting RecA binding, allowing RuvAB-dependent RFR. In the absence of UvrD, RarA and RecQJFOR promote RecA binding, preventing RFR. In the dnaNts mutant, formation of a RecA filament at blocked forks does not require RarA, and is not prevented by UvrD252. UvrD act only after RecA binding, presumably by removing RecA from DNA. The enzymes that catalyze RFR in this mutant are unknown. We propose that the different processing of forks in the two mutants results from differences in the disassembly of the replisome, symbolized by ovals of different colors. Small circles represent RecA.

The anti-RecA action of UvrD is conserved in B. subtilis

UvrD shares 42% sequence identity with PcrA and 37% with Rep. The three proteins also share very similar structures, even if they may use different modes of action (Korolev et al, 1997; Velankar et al, 1999; Lee and Yang, 2006). Rep is expressed in all mutants used here and is clearly unable to perform UvrD functions. This does not result from a lower level of proteins, since overexpression of Rep from a high-copy number plasmid did not restore the viability of Pol IIIts ΔuvrD mutants at 37°C (our unpublished data). It was previously reported that PcrA suppresses the lethality of rep uvrD double mutants, and the NER defect but not the mismatch repair defect of uvrD single mutants (Petit and Ehrlich, 2002). Furthermore, PcrA suppresses the lethality caused by uvrD inactivation in cells that carry additional Ter sites. Hence, PcrA shares with UvrD the capacity to remove the Tus protein from DNA in vivo (Bidnenko et al, 2006). Despite the homology between Rep, UvrD and PcrA, the ΔuvrD mutant defects are suppressed by the expression of PcrA and not Rep. This emphasizes the separation of function between Rep and UvrD and the conservation of UvrD functions in PcrA. We show here that PcrA also suppresses all defects of ΔuvrD mutants that are linked to RecA action: SOS induction, RecFOR-dependent hyper-recombination, lethality and lack of RFR in Pol IIIts contexts. These results are compatible with the idea that PcrA can remove E. coli RecA from DNA at gaps and at replication forks whereas Rep does not, in agreement with the observation that Rep is unable to undo E. coli RecA filaments in vitro (Veaute et al, 2005). If RecA removal requires an interaction between RecA and UvrD, this interaction is conserved in PcrA. RecA is highly conserved between E. coli and B. subtilis, and the conservation of the anti-RecA action of UvrD in a distant bacterial species underlines the importance of this function.

In conclusion, our results show that UvrD has multiple functions at inactivated replication forks, which are all linked to the action of recombination proteins but by different means. We previously reported that to remove DNA-bound Tus protein, UvrD acts in concert with RecBCD-dependent homologous recombination (Bidnenko et al, 2006). We show here that to rescue dnaNts-blocked forks, UvrD needs its ATPase activity and may remove RecA from DNA. In contrast, to allow replication restart from dnaEts-blocked forks, UvrD does not require either DNA translocation or unwinding activities. Therefore, UvrD is likely to act by preventing the access of RarA and RecQ either directly by steric hindrance, or indirectly by controlling the presence of other proteins at forks. This is the first indication that a protein of the UvrD family plays a role at replication forks that does not require any ATPase-dependent activity.

Materials and methods

Strains and plasmids

The strain background is JJC40, which is an hsdR Thr+ Pro+ derivative of AB1157 (leu-6 thi-1, his-4, argE3, lacY1, galK2, ara-14, xyl-5, mtl-1, tsx-33, rpsL31, supE44). Most of the strains were constructed by P1 transduction. Details of strain constructions and strains genotypes are described in in Supplementary Table S1. All thermosensitive mutants were constructed and propagated at 30°C, except for dnaEts ΔuvrD and dnaEts uvrD252 mutants and their derivatives, which were constructed and propagated at 25°C. recBCts stands for recB270 recC271 mutations (Supplementary Table S1). Because the viability of dnaEts (or dnaNts) recA recBCts uvrD252 mutants was very low, to measure chromosome breakage in a recA mutant context, RecBC was inactivated in the dnaEts (or dnaNts) recA uvrD252 mutants using a plasmid expressing the λ Gam protein under the control of the λ PL promoter, induced at 42 or 37°C. In previous works, where recBC could be inactivated either by a recBCts mutation or by Gam expression in a recA mutant context, we observed no difference between these two ways of inactivating recBC, as expected from the properties of the λ Gam protein (Grompone et al, 2002; Flores et al, 2005; Baharoglu et al, 2006). Null mutants were checked by PCR with external oligonucleotides that amplify a DNA fragment of different length for the wild-type and the interrupted alleles. Recombination mutations were checked as conferring UV sensitivity. In addition, recBC and recD mutants were checked for the inactivation of exonuclease V (they are permissive for the growth of T4 gpII mutants). UvrD mutants were checked for their UV sensitive and mutator phenotypes (increase in the proportion of RifR clones in overnight cultures). In PcrA+ strains, the B. subiltis pcrA gene is integrated in the lacZ gene of the E. coli chromosome under the control of a Pspac promoter. Pspac repression by LacI is weakly efficient and PcrA is expressed in our LacI+ background regardless of the presence of IPTG, as measured by the UV resistance conferred by the lacZ::PcrA insertion in ΔuvrD mutants.

Growth curves and measures of SOS induction

Overnight cultures grown at 30°C were diluted to OD 0.001 or 0.002 in Luria Broth medium (LB), grown at 30°C for two hours and then shifted to 37°C. Aliquots were taken and dilutions were plated on LB every hour. Plates were incubated for 48 h at 30°C. β-Galactosidase assays for the measure of SOS induction were performed as described (Miller, 1992).

Measure of linear DNA by PFGE

Quantification of pulsed field gels was performed as previously described (Seigneur et al, 1998). Values are considered as highly significantly different when P<0.001.

Supplementary Material

Supplementary data

7601804s1.doc (345KB, doc)

Acknowledgments

We are highly grateful to Marion Velten and Patrice Polard for their help in the realization of this work, their useful suggestions, constant interest and support. We thank Steven Mac Govern and Nicolas Sanchez for excellent technical assistance. We also thank Dr Marie-Agnès Petit, Dr Patrice Polard and Dr Steve Sandler for very helpful reading the manuscript. This work was supported by the Institut National de la Recherche Agronomique, by the ACI ‘Microbiologie 2003', the ACI ‘Biologie Cellulaire, Moléculaire et Structurale 2004' of the Ministère de la Recherche Française, by the FRM grant INE20051105275 and by the ANR grant ANR-05-BLAN-0204-01.

References

  1. Baharoglu Z, Petranovic M, Flores MJ, Michel B (2006) RuvAB is essential for replication forks reversal in certain replication mutants. EMBO J 25: 596–604 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Barre FX, Soballe B, Michel B, Aroyo M, Robertson M, Sherratt D (2001) Circles: the replication–recombination–chromosome segregation connection. Proc Natl Acad Sci USA 98: 8189–8195 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bidnenko V, Lestini R, Michel B (2006) The Escherichia coli UvrD helicase is essetial for Tus removal during recombination-dependent restart from Ter sites. Mol Microbiol 62: 382–396 [DOI] [PubMed] [Google Scholar]
  4. Bierne H, Seigneur M, Ehrlich SD, Michel B (1997) uvrD mutations enhance tandem repeat deletion in the Escherichia coli chromosome via SOS induction of the RecF recombination pathway. Mol Microbiol 26: 557–567 [DOI] [PubMed] [Google Scholar]
  5. Bruand C, Ehrlich SD (2000) UvrD-dependent replication of rolling-circle plasmids in Escherichia coli. Mol Microbiol 35: 204–210 [DOI] [PubMed] [Google Scholar]
  6. Centore RC, Sandler SJ (2007) UvrD limits the number and intensities of RecA-GFP structures in Escherichia coli K12. J Bacteriol 189: 2915–2920 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Fabre F, Chan A, Heyer WD, Gangloff S (2002) Alternate pathways involving Sgs1/Top3, Mus81/Mms4, and Srs2 prevent formation of toxic recombination intermediates from single-stranded gaps created by DNA replication. Proc Natl Acad Sci USA 99: 16887–16892 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Flores MJ, Bidnenko V, Michel B (2004) The DNA repair helicase UvrD is essential for replication fork reversal in replication mutants. EMBO Rep 5: 983–988 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Flores MJ, Sanchez N, Michel B (2005) A fork-clearing role for UvrD. Mol Microbiol 57: 1664–1675 [DOI] [PubMed] [Google Scholar]
  10. Gorbalenya AE, Koonin EV (1993) Helicases: amino acid sequence comparison and structure-function relationship. FEBS Lett 235: 16–242841153 [Google Scholar]
  11. Grompone G, Seigneur M, Ehrlich SD, Michel B (2002) Replication fork reversal in DNA polymerase III mutants of Escherichia coli: a role for the Beta clamp. Mol Microbiol 44: 1331–1339 [DOI] [PubMed] [Google Scholar]
  12. Guarne A, Ramon-Maiques S, Wolff EM, Ghirlando R, Hu X, Miller JH, Yang W (2004) Structure of the MutL C-terminal domain: a model of intact MutL and its roles in mismatch repair. EMBO J 23: 4134–4145 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hall MC, Jordan JR, Matson SW (1998) Evidence for a physical interaction between the Escherichia coli methyl-directed mismatch repair proteins MutL and UvrD. EMBO J 17: 1535–1541 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hishida T, Han YW, Shibata T, Kubota Y, Ishino Y, Iwasaki H, Shinagawa H (2004) Role of the Escherichia coli RecQ DNA helicase in SOS signaling and genome stabilization at stalled replication forks. Genes Dev 18: 1886–1897 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hishida T, Ohno T, Iwasaki H, Shinagawa H (2002) Saccharomyces cerevisiae MGS1 is essential in strains deficient in the RAD6-dependent DNA damage tolerance pathway. EMBO J 21: 2019–2029 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hishida T, Ohya T, Kubota Y, Kamada Y, Shinagawa H (2006) Functional and physical interaction of yeast Mgs1 with PCNA: impact on RAD6-dependent DNA damage tolerance. Mol Cell Biol 26: 5509–5517 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Horii Z, Clark AJ (1973) Genetic analysis of the recF pathway to genetic recombination in Escherichia coli K12: isolation and characterization of mutants. J Mol Biol 80: 327–344 [DOI] [PubMed] [Google Scholar]
  18. Johnson A, O'Donnell M (2005) Cellular DNA replicases: components and dynamics at the replication fork. Annu Rev Biochem 74: 283–315 [DOI] [PubMed] [Google Scholar]
  19. Kawabe Y, Branzei D, Hayashi T, Suzuki H, Masuko T, Onoda F, Heo SJ, Ikeda H, Shimamoto A, Furuichi Y, Seki M, Enomoto T (2001) A novel protein interacts with the Werner's syndrome gene product physically and functionally. J Biol Chem 276: 20364–20369 [DOI] [PubMed] [Google Scholar]
  20. Kolodner RD, Putnam CD, Myung K (2002) Maintenance of genome stability in Saccharomyces cerevisiae. Science 297: 552–557 [DOI] [PubMed] [Google Scholar]
  21. Korolev S, Hsieh J, Gauss GH, Lohman TM, Waksman G (1997) Major domain swiveling revealed by the crystal structures of complexes of E. coli Rep helicase bound to single-stranded DNA and ADP. Cell 90: 635–647 [DOI] [PubMed] [Google Scholar]
  22. Kuzminov A (1999) Recombinational repair of DNA damage in Escherichia coli and bacteriophage lambda. Microbiol Mol Biol Rev 63: 751–813 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lahue RS, Au KG, Modrich P (1989) DNA mismatch correction in a defined system. Science 245: 160–164 [DOI] [PubMed] [Google Scholar]
  24. Langston LD, O'Donnell M (2006) DNA replication: keep moving and don't mind the gap. Mol Cell 23: 155–160 [DOI] [PubMed] [Google Scholar]
  25. Lee JY, Yang W (2006) UvrD helicase unwinds DNA one base pair at a time by a two-part power stroke. Cell 127: 1349–1360 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Magner DB, Blankstein MD, Lee JA, Pennington JM, Lupski JR, Rosenberg SM (2007) RecQ promotes toxinc recombination in cells lacking recombination intermediate-removal proteins. Mol Cell 26: 273–286 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Matson SW, Robertson AB (2006) The UvrD helicase and its modulation by the mismatch repair protein MutL. Nucleic Acids Res 34: 4089–4097 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Michel B (2000) Replication fork arrest and DNA recombination. Trends Biochem Sci 25: 173–178 [DOI] [PubMed] [Google Scholar]
  29. Michel B, Grompone G, Flores MJ, Bidnenko V (2004) Multiple pathways process stalled replication forks. Proc Natl Acad Sci USA 101: 12783–12788 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Miller JH (1992) A Short Course in Bacterial Genetic. Cold Spring Harbor, New York: Cold Spring Harbor Press [Google Scholar]
  31. Morel P, Hejna JA, Ehrlich SD, Cassuto E (1993) Antipairing and strand transferase activities of E. coli Helicase-II (UvrD). Nucleic Acids Res 21: 3205–3209 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Morimatsu K, Kowalczykowski SC (2003) RecFOR proteins load RecA protein onto gapped DNA to accelerate DNA strand exchange: a universal step of recombinational repair. Mol Cell 11: 1337–1347 [DOI] [PubMed] [Google Scholar]
  33. Nakayama K, Irino N, Nakayama H (1985) The recQ gene of Escherichia coli K12: molecular cloning and isolation of insertion mutants. Mol Gen Genet 200: 266–271 [DOI] [PubMed] [Google Scholar]
  34. O'Donnell M (2006) Replisome architecture and dynamics in Escherichia coli. J Biol Chem 281: 10653–10656 [DOI] [PubMed] [Google Scholar]
  35. Ooi SL, Shoemaker DD, Boeke JD (2003) DNA helicase gene interaction network defined using synthetic lethality analyzed by microarray. Nat Genet 35: 277–286 [DOI] [PubMed] [Google Scholar]
  36. Ossanna N, Mount DW (1989) Mutations in uvrD induce the SOS response in Escherichia coli. J Bacteriol 171: 303–307 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Petit MA, Dervyn E, Rose M, Entian KD, McGovern S, Ehrlich SD, Bruand C (1998) PcrA is an essential DNA helicase of Bacillus subtilis fulfilling functions both in repair and rolling-circle replication. Mol Microbiol 29: 261–273 [DOI] [PubMed] [Google Scholar]
  38. Petit MA, Ehrlich D (2002) Essential bacterial helicases that counteract the toxicity of recombination proteins. EMBO J 21: 3137–3147 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Pfander B, Moldovan GL, Sacher M, Hoege C, Jentsch S (2005) SUMO-modified PCNA recruits Srs2 to prevent recombination during S phase. Nature 436: 428–433 [DOI] [PubMed] [Google Scholar]
  40. SaiSree L, Reddy M, Gowrishankar J (2000) Lon incompatibility associated with mutations causing SOS induction: null uvrD alleles induce the SOS response in Escherichia coli. J Bacteriol 182: 3151–3157 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Sandler SJ, Marians KJ (2000) Role of PriA in replication fork reactivation in Escherichia coli. J Bacteriol 182: 9–13 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Schmidt KH, Kolodner RD (2006) Suppression of spontaneous genome rearrangements in yeast DNA helicase mutants. Proc Natl Acad Sci USA 103: 18196–18201 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Seigneur M, Bidnenko V, Ehrlich SD, Michel B (1998) RuvAB acts at arrested replication forks. Cell 95: 419–430 [DOI] [PubMed] [Google Scholar]
  44. Seigneur M, Ehrlich SD, Michel B (2000) RuvABC-dependent double-strand breaks in dnaBts mutants require RecA. Mol Microbiol 38: 565–574 [DOI] [PubMed] [Google Scholar]
  45. Sherratt DJ, Soballe B, Barre FX, Filipe S, Lau I, Massey T, Yates J (2004) Recombination and chromosome segregation. Philos Trans R Soc Lond B Biol Sci 359: 61–69 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Shibata T, Hishida T, Kubota Y, Han YW, Iwasaki H, Shinagawa H (2005) Functional overlap between RecA and MgsA (RarA) in the rescue of stalled replication forks in Escherichia coli. Genes Cells 10: 181–191 [DOI] [PubMed] [Google Scholar]
  47. Shima N, Alcaraz A, Liachko I, Buske TR, Andrews CA, Munroe RJ, Hartford SA, Tye BK, Schimenti JC (2007) A viable allele of Mcm4 causes chromosome instability and mammary adenocarcinomas in mice. Nat Genet 39: 93–98 [DOI] [PubMed] [Google Scholar]
  48. Singleton MR, Dillingham MS, Gaudier M, Kowalczykowski SC, Wigley DB (2004) Crystal structure of RecBCD enzyme reveals a machine for processing DNA breaks. Nature 432: 187–193 [DOI] [PubMed] [Google Scholar]
  49. Veaute X, Delmas S, Selva M, Jeusset J, Le Cam E, Matic I, Fabre F, Petit MA (2005) UvrD helicase, unlike Rep helicase, dismantles RecA nucleoprotein filaments in Escherichia coli. EMBO J 24: 180–189 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Velankar SS, Soultanas P, Dillingham MS, Subramanya HS, Wigley DB (1999) Crystal structures of complexes of PcrA DNA helicase with a DNA substrate indicate an inchworm mechanism. Cell 97: 75–84 [DOI] [PubMed] [Google Scholar]
  51. Vijeh Motlagh ND, Seki M, Branzei D, Enomoto T (2006) Mgs1 and Rad18/Rad5/Mms2 are required for survival of Saccharomyces cerevisiae mutants with novel temperature/cold sensitive alleles of the DNA polymerase delta subunit, Pol31. DNA Repair (Amst) 5: 1459–1474 [DOI] [PubMed] [Google Scholar]
  52. Washburn BK, Kushner SR (1993) Characterization of DNA helicase-II from a uvrD252 mutant of Escherichia coli. J Bacteriol 175: 341–350 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. West SC (1997) Processing of recombination intermediates by the RuvABC proteins. Annu Rev Genet 31: 213–244 [DOI] [PubMed] [Google Scholar]
  54. Yamaguchi M, Dao V, Modrich P (1998) MutS and MutL activate DNA helicase II in a mismatch-dependent manner. J Biol Chem 273: 9197–9201 [DOI] [PubMed] [Google Scholar]
  55. Zieg J, Maples VF, Kushner SR (1978) Recombinant levels of Escherichia coli K-12 mutants deficient in various replication, recombination, or repair genes. J Bacteriol 134: 958–966 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary data

7601804s1.doc (345KB, doc)

Articles from The EMBO Journal are provided here courtesy of Nature Publishing Group

RESOURCES