Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Mol Cell. Author manuscript; available in PMC 2010 October 9.
Published in final edited form as:
PMCID: PMC2761248

An SOS Inhibitor that Binds to Free RecA protein: the PsiB protein


The process of bacterial conjugation involves the transfer of a conjugative plasmid as a single strand. The potentially deleterious SOS response, normally triggered by the appearance of single-stranded DNA, is suppressed in the recipient cell by a conjugative plasmid system centered on the product of the psiB gene. The F plasmid PsiB protein inhibits all activities of the RecA protein, including DNA binding, DNA strand exchange, and LexA protein cleavage. The proteins known to negatively regulate recombinases such as RecA or Rad51 generally work at the level of dismantling the nucleoprotein filament. However, PsiB binds to RecA protein that is free in solution. The RecA-PsiB complex impedes formation of RecA nucleoprotein filaments on DNA.


Conjugation is a mechanism of horizontal gene transfer that allows asexually reproducing bacteria to diversify their genomes. Conjugative plasmids aid the spread of antibiotic resistance in bacterial strains ((Aguero et al., 1984) and references therein), making the process of conjugation an important target for fighting disease (Lujan et al., 2007). In Escherichia coli, conjugation occurs between a donor cell harboring a conjugative plasmid and a recipient cell that lacks a similar plasmid (Frost et al., 1994). During conjugation, rolling circle replication ensues, and a single strand of the plasmid is spooled into the recipient ((Wolkow et al., 1996) and references therein).

The single-stranded DNA that enters the recipient cell can trigger the SOS response, involving induction of DNA repair genes that are under the control of the LexA transcription repressor protein (Roca and Cox, 1997; Walker et al., 2000). The LexA protein undergoes autocatalytic cleavage stimulated by RecA protein filaments bound to ssDNA and ATP (Little, 1991; Little et al., 1980; Roca and Cox, 1997; Slilaty and Little, 1987). During SOS, cell division halts, (Walker et al., 2000), and the induction of the mutagenic DNA polymerase V is possible if the response is not abated (Sommer et al., 1993; Walker et al., 2000). Therefore, the induction of SOS during conjugation is potentially detrimental to the cell.

Some plasmids can suppress unwarranted SOS induction during conjugation. The Plasmid SOS Interference/Inhibition (psi) phenomenon was first observed when the conjugative plasmid R100.1 suppressed the temperature-sensitive constitutive SOS phenotype of recA441 (Bagdasarian et al., 1980). Further studies with the R6-5 plasmid showed that a region encoding the PsiB protein alone was sufficient to suppress the SOS response (Bailone et al., 1988).

PsiB expression varies as a function of the promoters present on a given conjugative plasmid, and the cellular circumstances. The native F plasmid psiB promoter appears to be activated only during conjugation and only in the recipient (Bagdasarian et al., 1992). Early transcription of PsiB is likely mediated by an element of secondary structure in the incoming single strand that mimics an RNA polymerase promoter recognized by the Frpo sigma factor (Masai and Arai, 1997).

The strength of PsiB inhibition of SOS depends on the RecA allele and on the concentration of PsiB in the host cell. Constitutive SOS at high temperatures is suppressed in recA441 (tif-1) cells harboring the R100.1 plasmid. However, the SOS response was still induced in response to DNA damage. The RecA441 mutant protein binds better than WT RecA to ssDNA coated by single-stranded DNA binding protein (SSB) at high temperatures (Lavery and Kowalczykowski, 1990), and this characteristic may be significant in overcoming the inhibition by PsiB protein. In a separate experiment, plasmids derived from the R6-5 plasmid, containing a psi region, were able to suppress induction of wild type recA gene after DNA damage (Bagdasarian et al., 1986). Thus, the PsiB protein is less effective in suppressing SOS induction by RecA441 than by wild type RecA (Bagdasarian et al., 1980). The effect of PsiB is also dose-dependent (Bailone et al., 1988).

Biochemical characterization of the psiB gene product has been very limited. PsiB does not inhibit SOS in LexA deficient cells, ruling out the possibility that it is a transcription repressor (Bagdasarian et al., 1986). Based on lack of sequence homology it has been argued that PsiB is unlikely to act as a LexA mimic, competing directly for the RecA nucleoprotein filament necessary for LexA cleavage (Bailone et al., 1988). Other proposals include a direct PsiB inhibition of RecA function by competing for DNA binding sites, stabilizing a protein that binds to DNA and competes with RecA (e.g., SSB), or directly binding to RecA and preventing it from binding to DNA (Bagdasarian et al., 1986; Bailone et al., 1988). Based on all results, particularly on the unpublished effects of certain RecA mutant proteins, Devoret and colleagues favored an inhibition mediated by a direct interaction between PsiB and RecA in solution (Bailone et al., 1988). However, biochemical data relevant to the PsiB mechanism have not appeared, and a direct interaction between PsiB and RecA has never been demonstrated.

The direct interaction of RecA and PsiB would represent a previously undescribed mechanism for the regulation of RecA and related recombinases. The strict requirements for targeted recombination in eukaryotic cells have led to a search for eukaryotic recombinase modulators, with a focus on modulators of yeast and human RecA homologs. These efforts have revealed a network of regulators, including BRCA2, p53, Rad52, Rad54, Rad55/57, Rad54b, TGFβ1, the Srs2 helicase, and c-Abl (Busygina et al., 2008; Kanamoto et al., 2002; Krejci et al., 2001; Marmorstein et al., 1998; Tanaka et al., 2000; Yuan et al., 1998). None of these modulators is known to exert its primary effect on Rad51 protein by forming a complex with it in the absence of DNA. Modulators of the bacterial RecA protein have also drawn more attention recently. Proteins that aid, direct and stabilize RecA binding include RecBCD, RecFOR, and DinI (Anderson and Kowalczykowski, 1997; Lusetti et al., 2004b; Morimatsu and Kowalczykowski, 2003; Sakai and Cox, 2009). Negative regulation of RecA activity, mediated by the UvrD and RecX proteins, is also important (Drees et al., 2006; Veaute et al., 2005). As in the case of the eukaryotic recombinase modulators, none of these regulators exert their primary effects by forming a complex with RecA in the absence of DNA. Phosphorylation prevents the human Rad51 protein from catalyzing DNA strand exchange (Yuan et al., 1998), but RecA is not a kinase substrate (Marcandier et al., 1994).

Mechanisms of RecA inhibition can guide the search for undiscovered mechanisms of recombinase modulation in eukaryotes. In this paper we demonstrate that the PsiB protein from the F conjugative plasmid binds directly to RecA protein that is free in solution, forming a complex that inhibits RecA filament formation on SSB-coated ssDNA.


We used several methods to characterize PsiB activity. Throughout, PsiB concentrations are reported both in terms of monomers and dimers. Analytical ultracentrifugation showed a homogeneous population of dimers (V. Petrova, unpublished data). It is possible that each subunit has an independent interaction site, which makes the concentration of PsiB monomers potentially relevant. In many experiments, SSB was bound to the ssDNA prior to RecA protein addition. As we show below, PsiB appears to inhibit RecA filament formation on SSB-coated ssDNA.

PsiB inhibits RecA filament formation on circular single-stranded DNA

ATP hydrolysis by the RecA protein is highly DNA-dependent (Cox, 2003). This property of the RecA protein allows an indirect measurement of RecA binding to DNA by following the rate of ATP hydrolysis in a coupled spectrophotometric assay (Drees et al., 2004; Lindsley and Cox, 1990). To assess how PsiB affects RecA-mediated ATP hydrolysis in the presence of M13mp18 circular single-stranded DNA bound by SSB protein, PsiB was titrated into a reaction containing 2 μM RecA and 3 μM ssDNA (1 μM available DNA binding sites at 3 nucleotides/site). The rate of ATP hydrolysis of RecA was inhibited proportionally to the concentration of PsiB (Figure 1A), and almost completely abolished in the presence of 18 μM PsiB monomers or 9 μM PsiB dimers.

Figure 1
PsiB inhibits RecA filament formation on SSB-coated M13mp18 cssDNA

Fewer RecA filaments form on SSB-coated DNA in the presence of PsiB

The inhibition of ATP hydrolysis documented in Figure 1A could reflect either an inhibition of RecA filament formation or a reduction in the function of formed filaments. We visualized the RecA filaments formed in the presence and absence of PsiB to determine if PsiB affected the amounts of RecA protein bound to DNA. The results are shown in Figure 1, panels B and C. Under the conditions used, the rate of ATP hydrolysis is reduced by about 90% in the presence of the PsiB. By electron microscopy, the number of countable molecules in the presence of PsiB declined to approximately 70% of the number of molecules in the control lacking PsiB. Numbers of shorter/gapped filaments increased. Numbers of DNA molecules coated only with SSB more than doubled. However, a significant portion (11–27%) of the molecules were twisted in a manner that precluded precise length measurements (Figure 2B). Although we cannot relate the observed amount of bound RecA to the ATPase results quantitatively, the EM results qualitatively support a mechanism in which PsiB prevents RecA from binding to DNA. A catalogue of molecules observed in the electron microscopy is provided in Figure 1C.

Figure 2
PsiB inhibits RecA co-protease activity on LexA

PsiB inhibits RecA-catalyzed LexA autocleavage

Figure 1A demonstrates an almost complete inhibition of RecA filament formation on SSB-bound cssDNA in the presence of a four-fold excess of PsiB dimers relative to RecA monomers. With 15000 monomers per cell under normal growth conditions (Stohl et al., 2003), it is likely that RecA will be present in excess of PsiB dimers. In E. coli strains harboring the R6-5 plasmid 6000 monomers of PsiB per cell have been reported (Bagdasarian et al., 1992). We examined the effects of PsiB on LexA cleavage in vitro, with about two PsiB dimers per seven RecA monomers. Under the conditions tested (including 7 μM RecA protein and 9 μM ssDNA), the LexA autocleavage was complete approximately 60 min after the start of the reaction in the absence of PsiB. In contrast, when 1.6 μM PsiB dimers were present, little LexA cleavage was evident after 90 min (Figure 2). Note that previous studies have documented a more robust RecA-mediated cleavage of LexA than is observed in the control experiment here. The reduced rate is due to order of addition of reaction components in this experiment, with SSB preceding RecA and impeding RecA filament formation relative to the optimal conditions (RecA preceding SSB) used in earlier studies (e.g., (Gruenig et al., 2008)).

PsiB moderately inhibits RecA-catalyzed 3-strand exchange

PsiB delayed nicked circular product (ncproduct) formation in a reaction in which RecA catalyzed strand exchange between cssDNA initially bound by SSB and a homologous ldsDNA molecule (Figure 3). The rate and extent of ncproduct formation catalyzed by RecA depended on the order of addition of SSB and RecA to DNA. When RecA was allowed to nucleate on cssDNA first, and SSB was added second, there was robust joint molecule (jm) and ncproduct formation 15 min after the reaction was started with the addition of ldsDNA. However, when SSB was allowed to coat the cssDNA first, before RecA was added, the reaction was slowed. When PsiB was present in the reaction in which RecA had to nucleate on SSB-coated DNA, product formation was further delayed in both rate and extent. The ratios of DNA to RecA protein to SSB protein were similar to that in the ATPase experiments above. However, the total protein and DNA concentrations were greater in the strand exchange experiments.

Figure 3
PsiB inhibits RecA-catalyzed recombination of homologous DNA in vitro

The effects of PsiB are diminished in the absence of SSB

The inhibition of RecA filament formation on DNA by PsiB may be due to PsiB alone or PsiB association with an interaction partner. To begin to examine the potential interaction partners of PsiB that are necessary to achieve an inhibition of RecA binding to DNA, we omitted the SSB protein from the ATPase reactions. The SSB protein is necessary to melt secondary structure in cssDNA (Kowalczykowski and Krupp, 1987), which enables RecA to bind along the entire length of the DNA molecule. Therefore, we used linear poly(dT) DNA for this set of experiments. Again, PsiB inhibited the RecA ATPase activity, reflecting a reduction of RecA binding to DNA (Figure 4A). However, the effect saturated before complete inhibition occurred. The ATPase activity of RecA did not fall much below 50% of the rate observed in the control reaction without PsiB (Figure 4B) even in the presence of 33.7 μM PsiB dimers.

Figure 4
PsiB partially inhibits RecA ATP hydrolysis on poly(dT)

The effects of SSB on the scope of inhibition by PsiB (compare Figure 1A to Figure 4A) hint at further mechanistic complexity. In order to understand the role of SSB in the inhibition of RecA by PsiB, we decided to use the DinI protein as a tool. DinI stabilizes RecA nucleoprotein filaments (Lusetti et al., 2004b) and counteracts the effect of the RecX protein, another inhibitor of RecA (Lusetti et al., 2004a). We wanted to examine how the stabilization of the RecA filaments affects the ability of PsiB to inhibit RecA.

When DinI was added prior to RecA and at the same time as PsiB to ATPase reactions containing poly(dT) DNA and no SSB, DinI completely counteracted the inhibitory effect of PsiB (Figure 5A). If preventing RecA from binding to DNA had been the only way PsiB inhibited RecA, then the presence or absence of SSB should not be an issue. However, that was not the case. Figure 5B illustrates that when SSB was present, RecA inhibition by PsiB on M13mp18 DNA was diminished in the presence of DinI but not completely abolished. This suggests that SSB has a significant role in the PsiB inhibition (see Discussion).

Figure 5
DinI counteracts the PsiB effect in the absence, but not in the presence of SSB

We investigated the dynamics of the PsiB-RecA interaction by testing whether DinI can disturb the inhibition of RecA by PsiB after the PsiB-RecA complexes are formed. RecA was incubated with PsiB and poly(dT) DNA, and after 15 min DinI was added, and the change in ATP hydrolysis was monitored. A restoration of the ATP hydrolysis was observed, indicating that the PsiB-RecA interaction was readily reversible (Figure 5C). In contrast, when a filament stabilized by DinI was challenged with PsiB, little effect was seen. This again indicates that PsiB has little effect on RecA protein once it is bound to ssDNA.

The PsiB protein does not interact with the LexA or SSB proteins, or with linear ssDNA

FITC-labeled PsiB protein was titrated with LexA protein, and the potential interaction between the two proteins was examined using fluorescence anisotropy. No evidence for interaction was observed using LexA protein concentrations up to 5 μM (Figure 6A).

Figure 6
PsiB interactions with LexA, SSB, DNA and RecA

Since SSB affects how strongly PsiB inhibits RecA, PsiB and SSB may interact. We investigated the potential existence of direct interactions between the two proteins by fluorescence anisotropy. FITC-labeled PsiB protein was titrated with SSB protein. Figure 6B illustrates that PsiB did not interact with SSB in solution, since no significant change in PsiB anisotropy occurs even when SSB is present at a 100 fold molar excess. The absence of interaction between the two proteins obtained by fluorescence anisotropy was confirmed in an electrophoretic mobility shift assay (EMSA), monitoring SSB and PsiB in the presence DNA (Figure 6C). Therefore, PsiB did not interact with SSB either in solution or when SSB was bound to DNA under the conditions of these experiments.

Even though it has been reported that PsiB does not bind to ssDNA (Bailone et al., 1988), we decided to test whether an interaction between PsiB and ssDNA exists in a more sensitive and quantifiable experiment. Fluorescence anisotropy measurements showed that the native PsiB protein did not interact with a FITC-labeled 50 nt linear single-stranded DNA oligomer even when PsiB is present at 103 excess relative to DNA molecules (Figure 6D). SSB was used as a positive control because of its high DNA-binding activity.

PsiB interacts with the RecA protein free in solution

One of the mechanisms proposed for PsiB is a direct interaction between PsiB and RecA when the latter protein is not bound to DNA (Bailone et al., 1988). We tested for this interaction in a fluorescence polarization experiment by incubating 60 nM labeled PsiB with increasing concentrations of the RecA protein. There was a large change in anisotropy of the probe (Figure 6E), indicating a PsiB-RecA interaction. The anisotropy continued increasing as more RecA protomers were added, but did not achieve saturation in the range of concentrations tested.

To show that the continued increase in anisotropy was due to a specific RecA-PsiB interaction and not to a contribution from the intrinsic fluorescence of RecA protein, we labeled BSA with FITC dye. The labeled BSA was diluted until its signal was of the same intensity as PsiB and was similarly titrated with increasing concentrations of RecA. No changes in the BSA signal were observed, indicating that the RecA protein was not contributing a background signal.

Since saturation of the anisotropy signal was not achieved, the dissociation constant could not be calculated. In addition, only 1 out of 11 PsiB monomers in this assay were labeled, possibly suppressing a signal increase at lower RecA concentrations. The observed increase begins in the nanomolar range of RecA concentration, indicating that a Kd in the low μM range is possible. Later testing of a more heavily labeled preparation of PsiB indicated that FITC labeling modestly increases the RecA inhibition activity of PsiB (data not shown).


We conclude that the PsiB protein inhibits RecA filament formation on SSB-coated ssDNA. That inhibition is sufficient to prevent the RecA-catalyzed LexA autocleavage necessary to induce the SOS response. We propose that the inhibition occurs via a mechanism not previously observed for a RecA family – recombinase an interaction between the RecA and PsiB proteins in solution. This work provides biochemical evidence for a mechanism proposed by Devoret and colleagues in their studies two decades ago (Bagdasarian et al., 1986; Bailone et al., 1988). The inhibition mediated by PsiB is much greater if the ssDNA to which RecA is to be bound is coated with SSB. The addition of PsiB to a RecA reaction reduces the rate and extent of RecA protein filament formation on SSB-coated ssDNA as measured by both the indirect RecA-mediated ATPase assay (Fig. 1A) and electron microscopy (Fig. 1B,C). The PsiB protein interacts with free RecA protein (Figure 6E), but does not interact measurably with SSB, LexA, or DNA (Figure 6A–D).

In suppressing the SOS response, the key effect of PsiB is the resulting inhibition of LexA autocleavage. Little LexA cleavage was seen for 90 min even when RecA monomers were in over four-fold excess relative to PsiB dimers (Fig. 2). PsiB inhibits RecA-mediated SOS induction in vivo in spite of being present at lower concentrations in the cell (3000 PsiB dimers vs 15,000 RecA monomers, (Bagdasarian et al., 1992; Stohl et al., 2003)). The observed in vitro inhibition of RecA-facilitated autocatalytic LexA cleavage is thus sufficient to explain the suppression of the SOS response in vivo caused by PsiB during conjugation. The inhibition of LexA autocatalytic cleavage by PsiB is the strongest effect we observed for PsiB, a level of inhibition that seems to eclipse the inhibition of RecA filament formation. The protracted inhibition of LexA cleavage even after some RecA filaments are demonstrably established on the DNA may telegraph an additional mechanistic contribution of PsiB, such as a continued association of PsiB with the assembled RecA filament. If such a continued association occurs, it does not appear to affect ATP hydrolysis by the RecA filament, or DNA strand exchange. An inhibition of ATP hydrolysis is not needed, since ATP binding but not hydrolysis is required for RecA to form a nucleoprotein filament capable of stimulating LexA autocleavage (Gruenig et al., 2008; Little et al., 1980). This supports the observations in vivo that the main role of PsiB is to inhibit the induction of the SOS response. Our experiment also supports the suggestion of Sandler and colleagues that there is a cellular threshold in the number or quality of RecA nucleoprotein filaments required for SOS induction (Centore et al., 2008; Gruenig et al., 2008). Thus, PsiB does not have to completely block RecA filamentation; it simply needs to prevent the attainment of the threshold level required for SOS induction. The molecular basis of the putative threshold has not been elucidated.

In principle, PsiB could also inhibit LexA autocleavage by either mimicking LexA and competing for binding to the RecA nucleoprotein filament or binding to LexA to prevent it from binding to RecA filaments on DNA. Fluorescence polarization experiments provide no evidence that PsiB interacts with LexA. Therefore, the PsiB mechanism is adequately described by invoking the demonstrated PsiB interaction with RecA.

A direct interaction between PsiB and RecA was proposed by Devoret and colleagues two decades ago (Bagdasarian et al., 1986; Bailone et al., 1988). The strength of the PsiB inhibition effect on SOS varies depending on what RecA phenotype is displayed by the strain harboring the psiB gene (Bagdasarian et al., 1986; Bagdasarian et al., 1980; Bailone et al., 1988). This result can now be rationalized in the context of a biochemically established PsiB-RecA interaction. The dose-dependent effect of PsiB on suppressing SOS (Bailone et al., 1988) can also be explained by the PsiB-RecA interaction.

The experiments incorporating the RecA filament-stabilizing protein DinI reveal a dynamic interaction between PsiB and RecA. As shown on Figure 5C, DinI can stabilize RecA on poly(dT) DNA in the absence of SSB even when RecA has been pre-incubated with PsiB in the reaction. In contrast, the interaction between DinI and RecA bound to DNA is stable, so PsiB cannot easily disrupt the preformed DinI-RecA filament complexes.

In Figure 7, we present a model for PsiB function that takes into account the evident importance of SSB in the inhibition patterns. We propose that RecA possesses an SSB displacement site in close proximity to or overlapping with the RecA DNA-binding site. We define an SSB displacement site to be a region that either directly interacts with SSB or that recognizes the DNA strand wrapped around SSB. The existence of such a RecA-SSB interaction site is implied by the capacity of C-terminal truncation mutants of RecA to bypass the need for mediator proteins in filament formation on SSB-coated ssDNA (Eggler et al., 2003). In effect, RecA can readily displace SSB from ssDNA; however, that capacity is suppressed by the RecA C-terminus. The RecA C-terminus itself acts as a kind of regulatory flap, occluding the SSB-displacement surface under most conditions. In our model, PsiB binds RecA primarily so as to cover the SSB-displacement surface, leading to the observed restricted access to DNA. This model is supported by the observation that PsiB inhibits filament formation onto SSB-coated ssDNA by a RecA mutant protein lacking the C-terminus as much as it inhibits WT RecA, eliminating the advantage of C-terminal deletion (RecA ΔC17; V. Petrova, unpublished data). The binding of PsiB in the SSB-displacement site must also partially occlude the RecA binding site for DNA, thus, decreasing RecA affinity for DNA. This would explain the partial inhibition of the RecA ATPase activity on poly(dT) DNA when SSB is absent. The SSB-displacement surface that we postulate on the RecA protein has not yet been structurally defined.

Figure 7
Proposed mechanism of PsiB activity

Experimental Procedures

Chemicals were purchased from Fisher and Sigma, except for DTT, which was purchased from Research Organics, and IPTG from Gold Biotechnology. NdeI and BamHI and T4 ligase were purchased from New England BioLabs.

Protein purification

The psiB gene was cloned from JM101 E. coli strain, harboring an F′ conjugative plasmid. The gene was PCR-amplified, cloned into pET 21a vector, and transformed into the BL21 strain. Deviation from the published psiB gene sequence was at bases 409 and 410, where T and C, respectively, were inverted relative to the sequence posted at NCBI. The originally published sequence (Dutreix et al., 1988) agreed with the sequencing we observed. Harvested cells were lysed by sonication in 250 mM Tris-HCl (80% cation), 25% sucrose, EDTA. PsiB protein was precipitated from the clarified cell lysate by addition of ammonium sulfate to 43% saturation. The pellet was washed with 1 M ammonium sulfate twice and resuspended in 20 mM Tris, 80% cation, 1 mM EDTA, 10% glycerol (R buffer, pH 7.7), dialyzed into R buffer, and then further dialyzed into 20 mM phosphate, 1 mM EDTA, 10% glycerol (P buffer, pH 7.0). The protein was purified successively on ceramic hydroxyapatite, Source S and Source Q columns. The resulting PsiB was >95% pure by gel and nuclease free, except for the presence of a minor double-stranded DNA endonuclease. This contaminating activity relaxed less than 10% of 20 μM supercoiled plasmid DNA over a period of 2 hours under conditions typically used in our experiments. The analysis of the protein by MS gave a major peak at 15878 Da (predicted Mr = 15880 Da). The concentration of PsiB was determined using a native extinction coefficient 1.35 × 104 M−1 cm−1, determined by a published procedure (Edelhoch, 1967). A second preparation, found to be substantially nuclease free, was used for DNA strand exchange experiments and to repeat several of the reported experiments.

Wild type RecA and SSB proteins from E. coli were purified with minor modifications as described (Lusetti et al., 2003; Lohman and Overman, 1985). The RecA preparation included an additional size exclusion step. RecA concentration was determined as previously from the protein absorbance at 280 nm, corrected for scattering at 320 nm using extinction coefficient 2.23 × 10−4 M−1 cm−1. The SSB preparation included a heparin and DEAE columns. The E. coli SSB protein used in fluorescence polarization was a kind gift of James Keck. SSB concentration was determined from the protein absorbance at 280 nm, corrected for scattering at 320 nm using the extinction coefficient 2.83 × 10−4 M−1 cm−1. The E. coli DinI and LexA proteins were purified as previously described (Gruenig et al., 2008; Lusetti et al., 2004a). Concentrations of the DinI and LexA proteins were determined using the extinction coefficients 1.44 × 104 M−1 cm−1 and 7300 M−1 cm−1, respectively.

DNA substrates

M13mp18 css and ds DNA were purified as previously described (Messing, 1983; Neuendorf and Cox, 1986). Poly(dT) DNA was obtained from Amersham, lot GG0076. Fluorescent oligomer den7 5′-GGC CTC GCG GTA GCT GAG CTC GGA GCG CAC GAT TCG CAC TGC TGA TGT TC-3′-fluoresceine was obtained from IDT.

ATPase assay

The ATPase activity was followed using a coupled assay monitored in a Varian Cary 300 dual beam spectrophotometer, equipped with a temperature controller and a 12-position cell changer as previously described (Lindsley and Cox, 1989; Morrical et al., 1986). The path length and the band pass were 1 cm or 0.5 cm and 2 nm respectively. Regeneration of ATP from ADP in the presence of PEP was coupled to the oxidation of NADH, and the decrease in NADH concentration was followed at 380 nm. The extinction coefficient of NADH (1.21 mM−1 cm−1) at 380 nm was used to calculate the amount of ATP hydrolyzed.

The reactions were carried out in 25 mM Tris-OAc (80% cation), 1mM DTT, 3 mM potassium glutamate, 10 mM magnesium acetate, 5 % glycerol (RecA buffer), an ATP regeneration system (10 u/ml pyruvate kinase and 3.5 mM PEP), a coupling system (1.5 mM NADH and 10 u/ml lactate dehydrogenase) and 3 μM DNA. The Figure legends describe the DNA type used in the assays.

Where RecA protein modulators of function (i.e. PsiB, DinI) were present, the modulators were incubated with the master mix at 37°C for 10 min, and the reactions were started with the addition of RecA. The only exception is the PsiB/DinI challenge assay where the modulator proteins or their storage buffers are added to the reaction 15 min after the addition of 2 μM RecA. PsiB concentrations were varied in the assays, and are pointed out in the Figure legends.

Strand Exchange

Two orders of addition were used in this experiment. In one of the controls, RecA was allowed to nucleate on css M13mp18 DNA, and then ATP and SSB were added. Strand-exchange was initiated by the addition of ldsDNA. In the experiment and second control for PsiB inhibition of RecA strand-exchange activity on SSB-coated DNA, the following order of addition was chosen. The reaction containing css M13mp18 DNA in RecA buffer, an ATP regeneration system (10 u/ml pyruvate kinase and 2.5 mM PEP), 3 μM ATP and SSB was incubated for 10 min to allow SSB to bind to cssDNA. PsiB or compensatory PsiB storage buffer was added and the reaction was further incubated for 10 min. RecA protein was added to the reaction and allowed to bind to the SSB-coated cssDNA for 10 min. The reaction was initiated with the addition of homologous ldsDNA. Time point 0 represents the immediate withdrawal of a sample from the reaction after ldsDNA addition and mixing. The concentrations of the proteins and DNA used in the experiment are given in the figure legends.

Fluorescence Polarization

Pure PsiB protein was labeled by dialyzing it into 10 mM sodium bicarbonate, pH 9, 10% glycerol and incubating with molar excess of fluoresceine isothiocyanide (FITC) (Sigma F7250) dissolved in DMSO over night at 4°C (Lopper et al., 2007). The reaction was quenched with NH4Cl (Petrushenko et al., 2002) to a final concentration of 50 mM, followed by dialysis into 20 mM Tris-Cl, 80% cation, 50 mM KCl, 10% glycerol. The free dye was removed by centrifuging the labeled protein in microcon tubes three times, and adding fresh storage buffer every time. The degree of labeling was determined by measuring the absorbance at 280 and 494 nm. The concentration of FITC was determined using extinction coefficient 68000 M−1 cm−1 at 494 nm, and the protein concentration was determined by the following formula: dilution factor*(A280-0.3*A494)/PsiB extinction coefficient. The degree of labeling was one FITC label per 11 PsiB monomers. The low degree of labeling can be explained by the presence of only one lysine per monomer of PsiB. The protein was stored at −80°C. Testing of a subsequent, and more highly labeled preparation indicated that FITC labeling may modestly increase PsiB activity. The measurements were done with Beacon 2000 (Invitrogen). The occurrence or absence of binding was determined by following the change in anisotropy of the labeled PsiB.

PsiB-LexA interactions

LexA protein was diluted in 20 mM Tris, 80% cation, pH 7.5, 200 mM potassium glutamate, 50 mM KCl, 1 mM magnesium acetate, 1 mM ATP, 4 % glycerol, 0.1 g/l BSA, 1 mM DTT buffer. The dilutions were used to titrate 60 nM FITC-PsiB probe. The dilutions for each concentration were prepared twice.

PsiB-SSB interactions

The SSB protein was dialyzed into 20 mM Tris-Cl, 80% cation, 50 mM KCl, 10% glycerol, 1 mM DTT and supplemented with BSA to 0.1 g/l final concentration. Dilutions were made into the same buffer and used to titrate 60 nM labeled PsiB protein. The dilutions for each concentration were prepared three times. The intensity of the measurements was at least 10 times the intensity of the background blank.

PsiB-DNA interactions

The PsiB protein was diluted into 20 mM Tris, 80% cation, 50 mM NaCl, 4 % glycerol, 0.1 g/l BSA, 1 mM DTT. Ten nM of den 7 DNA was titrated with increasing concentrations of PsiB protein. SSB protein was used as a positive control for a protein of similar molecular weight that binds DNA. The highest dilution from the PsiB stock was from 1:10 (final concentration of 15.6 μM), which would have resulted in increase of glycerol concentration in the reaction from 4% to 4.6%. We assume that such increase will have a negligible effect on the potential interaction between PsiB and DNA at such high excess of protein.

PsiB-RecA interactions

The RecA-PsiB FP experiments were carried out as described for the PsiB-LexA FP experiments. The dilutions for each concentration were done 3–9 times. The horizontal and vertical intensities of the sample were at least four times the intensities of the blank. In the control with labeled BSA and RecA, BSA was diluted until it had the same intensity as the labeled PsiB used in the reactions.


The reactions were conducted in the same buffer as the ATPase assays. SSB (0.5 μM monomers) was incubated with den7 DNA for 10 min, and then incubated with PsiB (concentrations shown on figure) or RecQ (0.25μM) for another 10 min. The complexes were analyzed on 10% native polyacrylamide gel. The position of the DNA on the gel was visualized using Typhoon 9410 scanner (Amersham Biosciences) with 488 nm excitation and 520 emission laser settings.

LexA Cleavage Assay

The assay was done generally as described (Drees et al., 2004), albeit with a different order of addition. M13mp18 cssDNA was incubated with SSB and PsiB for 10 min at 37°C. RecA protein was added and allowed to equilibrate for 10 min. The reactions are initiated by the addition of LexA protein.

Electron Microscopy

Carbon films, mounted on a 400-mesh electron microscopy grids, were first activated by a brief glow-discharge treatment (Grassucci et al., 2007). Activated grids were used immediately. Samples for electron microscopy were prepared in RecA buffer. The experiment consisted of a 10 min pre-incubation of 0.5 μM SSB and either 9 μM PsiB protein or the appropriate volume of compensatory storage buffer with 3 μM M13mp18 cssDNA in the presence of 3 μM ATP, and an ATP regeneration system containing 12 mM phosphocreatine and 10 units of creatine kinase. The reaction was then initiated with 2 μM RecA and allowed to proceed for 100 min. All incubations were done at 37°C. The reaction mixture described above was diluted 4 times with RecA buffer, and 8 μl of the diluted sample was immediately adsorbed to the activated carbon film for 3 minutes. The grid was then touched to a drop of the above buffer followed by floating on a fresh drop of the same buffer for 1 min. The sample was stained by touching to a drop of 5% uranyl acetate followed by floating on a fresh drop of the same solution for 30 sec. Finally, the grid was washed by touching to a drop of double distilled water followed by immersion in two 10-ml beakers of double distilled water. After the sample was dried, it was rotary-shadowed with platinum. This protocol was designed for visualization of complete reaction mixtures, and no attempt was made to remove unreacted material. Although this approach should yield results that give a true insight into reaction components, it does lead to samples with a high background of unreacted proteins.

To determine the proportion of the molecules observed that were either fully or partially coated by RecA protein or bound only by the SSB protein, at least eight separate regions of the grids from four different experiments were counted at an identical magnification for each sample. A molecule was considered discontinuous, or gapped, if it had a detectable region of SSB-coated DNA of any size or there is a gap between two adjacent protein molecules. Imaging and photography were carried out with a TECNAI G2 12 Twin Electron Microscope (FEI Co.) equipped with a GATAN 890 CCD camera. Digital images of the nucleoprotein filaments were taken at X 15000 Magnification.


We would like to thank Liz Wood for cloning the psiB gene, Duo Lu and Nick George from the Keck lab for FP tutorials, Nick George for the kind gift of SSB, used in the EMSA experiments and Rob Shereda for the kind gift of RecQ. This work was supported by grant GM32335 to MMC from the Institute of General Medical Sciences of the United States National Institutes of Health.


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


  • Aguero ME, Aron L, DeLuca AG, Timmis KN, Cabello FC. A plasmid-encoded outer membrane protein, TraT, enhances resistance of Escherichia coli to phagocytosis. Infect Immun. 1984;46:740–746. [PMC free article] [PubMed]
  • Anderson DG, Kowalczykowski SC. The translocating RecBCD enzyme stimulates recombination by directing RecA protein onto ssDNA in a chi-regulated manner. Cell. 1997;90:77–86. [PubMed]
  • Bagdasarian M, Bailone A, Angulo JF, Scholz P, Devoret R. PsiB, an anti-SOS protein, is transiently expressed by the F sex factor during its transmission to an Escherichia coli K-12 recipient. Mol Microbiol. 1992;6:885–893. [PubMed]
  • Bagdasarian M, Bailone A, Bagdasarian MM, Manning PA, Lurz R, Timmis KN, Devoret R. An inhibitor of SOS induction, specified by a plasmid locus in Escherichia coli. Proc Natl Acad Sci USA. 1986;83:5723–5726. [PubMed]
  • Bagdasarian M, D’Ari R, Filipowicz W, George J. Suppression of induction of SOS functions in an Escherichia coli tif-1 mutant by plasmid R100.1. J Bacteriol. 1980;141:464–469. [PMC free article] [PubMed]
  • Bailone A, Backman A, Sommer S, Celerier J, Bagdasarian MM, Bagdasarian M, Devoret R. PsiB polypeptide prevents activation of RecA protein in Escherichia coli. Mol Gen Genet. 1988;214:389–395. [PubMed]
  • Busygina V, Sehorn MG, Shi IY, Tsubouchi H, Roeder GS, Sung P. Hed1 regulates Rad51-mediated recombination via a novel mechanism. Genes Develop. 2008;22:786–795. [PubMed]
  • Centore RC, Lestini R, Sandler SJ. XthA (Exonuclease III) regulates loading of RecA onto DNA substrates in log phase Escherichia coli cells. Mol Microbiol. 2008;67:88–101. [PubMed]
  • Cox MM. The bacterial RecA protein as a motor protein. Annu Rev Microbiol. 2003;57:551–577. [PubMed]
  • Drees JC, Chitteni-Pattu S, McCaslin DR, Inman RB, Cox MM. Inhibition of RecA protein function by the RdgC protein from Escherichia coli. J Biol Chem. 2006;281:4708–4717. [PubMed]
  • Drees JC, Lusetti SL, Chitteni-Pattu S, Inman RB, Cox MM. A RecA filament capping mechanism for RecX protein. Mol Cell. 2004;15:789–798. [PubMed]
  • Dutreix M, Backman A, Celerier J, Bagdasarian MM, Sommer S, Bailone A, Devoret R, Bagdasarian M. Identification of psiB genes of plasmids F and R6-5. Molecular basis for psiB enhanced expression in plasmid R6–5 Nuc. Acids Res. 1988;16:10669–10679. [PMC free article] [PubMed]
  • Edelhoch H. Spectroscopic determination of tryptophan and tyrosine in proteins. Biochemistry. 1967;6:1948–1954. [PubMed]
  • Eggler AL, Lusetti SL, Cox MM. The C terminus of the Escherichia coli RecA protein modulates the DNA binding competition with single-stranded DNA-binding protein. J Biol Chem. 2003;278:16389–16396. [PubMed]
  • Frost LS, Ippen-Ihler K, Skurray RA. Analysis of the sequence and gene products of the transfer region of the F sex factor. Microbiol Rev. 1994;58:162–210. [PMC free article] [PubMed]
  • Grassucci RA, Taylor DJ, Frank J. Preparation of macromolecular complexes for cryo-electron microscopy. Nature Prot. 2007;2:3239–3246. [PMC free article] [PubMed]
  • Gruenig MC, Renzette N, Long E, Chitteni-Pattu S, Inman RB, Cox MM, Sandler SJ. RecA-mediated SOS induction requires an extended filament conformation but no ATP hydrolysis. Mol Microbiol. 2008;69:1165–1179. [PMC free article] [PubMed]
  • Kanamoto T, Hellman U, Heldin CH, Souchelnytskyi S. Functional proteomics of transforming growth factor-beta1-stimulated Mv1Lu epithelial cells: Rad51 as a target of TGFbeta1-dependent regulation of DNA repair. EMBO J. 2002;21:1219–1230. [PubMed]
  • Kowalczykowski SC, Krupp RA. Effects of Escherichia coli SSB protein on the single-stranded DNA-dependent ATPase activity of Escherichia coli RecA protein. Evidence that SSB protein facilitates the binding of RecA protein to regions of secondary structure within single-stranded DNA. J Mol Biol. 1987;193:97–113. [PubMed]
  • Krejci L, Damborsky J, Thomsen B, Duno M, Bendixen C. Molecular dissection of interactions between Rad51 and members of the recombination-repair group. Mol Cell Biol. 2001;21:966–976. [PMC free article] [PubMed]
  • Lavery PE, Kowalczykowski SC. Properties of recA441 protein-catalyzed DNA strand exchange can be attributed to an enhanced ability to compete with SSB protein. J Biol Chem. 1990;265:4004–4010. [PubMed]
  • Lindsley JE, Cox MM. Dissociation pathway for RecA nucleoprotein filaments formed on linear duplex DNA. J Mol Biol. 1989;205:695–711. [PubMed]
  • Lindsley JE, Cox MM. Assembly and disassembly of RecA protein filaments occurs at opposite filament ends: relationship to DNA strand exchange. J Biol Chem. 1990;265:9043–9054. [PubMed]
  • Little JW. Mechanism of specific LexA cleavage - autodigestion and the role of RecA coprotease. Biochimie. 1991;73:411–422. [PubMed]
  • Little JW, Edmiston SH, Pacelli LZ, Mount DW. Cleavage of the Escherichia coli LexA protein by the RecA protease. Proc Natl Acad Sci USA. 1980;77:3225–3229. [PubMed]
  • Lohman TM, Overman LB. Two binding modes in Escherichia coli single strand binding protein-single stranded DNA complexes. Modulation by NaCl concentration. J Biol Chem. 1985;260:3594–3603. [PubMed]
  • Lopper M, Boonsombat R, Sandier SJ, Keck JL. A hand-off mechanism for primosome assembly in replication restart. Mol Cell. 2007;26:781–793. [PMC free article] [PubMed]
  • Lujan SA, Guogas LM, Ragonese H, Matson SW, Redinbo MR. Disrupting antibiotic resistance propagation by inhibiting the conjugative DNA relaxase. Proc Natl Acad Sci USA. 2007;104:12282–12287. [PubMed]
  • Lusetti SL, Drees JC, Stohl EA, Seifert HS, Cox MM. The DinI and RecX proteins are competing modulators of RecA function. J Biol Chem. 2004a;279:55073–55079. [PubMed]
  • Lusetti SL, Voloshin ON, Inman RB, Camerini-Otero RD, Cox MM. The DinI protein stabilizes RecA protein filaments. J Biol Chem. 2004b;279:30037–30046. [PubMed]
  • Lusetti SL, Wood EA, Fleming CD, Modica MJ, Korth J, Abbott L, Dwyer DW, Roca AI, Inman RB, Cox MM. C-terminal deletions of the Escherichia coli RecA protein - Characterization of in vivo and in vitro effects. J Biol Chem. 2003;278:16372–16380. [PubMed]
  • Marcandier S, Granger-Schnarr M, Cozzone AJ. Phosphorylation of Escherichia coli proteins during the SOS response. Int J Biochem. 1994;26:387–396. [PubMed]
  • Marmorstein LY, Ouchi T, Aaronson SA. The BRCA2 gene product functionally interacts with p53and RAD51. Proc Natl Acad Sci USA. 1998;95:13869–13874. [PubMed]
  • Masai H, Arai K. Frpo: a novel single-stranded DNA promoter for transcription and for primer RNA synthesis of DNA replication. Cell. 1997;89:897–907. [PubMed]
  • Messing J. New M13 vectors for cloning. Methods Enzymol. 1983;101:20–78. [PubMed]
  • Morimatsu K, Kowalczykowski SC. RecFOR proteins load RecA protein onto gapped DNA to accelerate DNA strand exchange: A universal step of recombinational repair. Mol Cell. 2003;11:1337–1347. [PubMed]
  • Morrical SW, Lee J, Cox MM. Continuous association of Escherichia coli single-stranded DNA binding protein with stable complexes of RecA protein and single-stranded DNA. Biochemistry. 1986;25:1482–1494. [PubMed]
  • Neuendorf SK, Cox MM. Exchange of RecA protein between adjacent RecA protein-single-stranded DNA complexes. J Biol Chem. 1986;261:8276–8282. [PubMed]
  • Petrushenko ZM, Budkevich TV, Shalak VF, Negrutskii BS, El’skaya AV. Novel complexes of mammalian translation elongation factor eEF1A.GDP with uncharged tRNA and aminoacyl-tRNA synthetase. Implications for tRNA channeling. Eur J Biochem. 2002;269:4811–4818. [PubMed]
  • Roca AI, Cox MM. RecA protein: structure, function, and role in recombinational DNA repair. Prog Nuc Acid Res Mol Biol. 1997;56:129–223. [PubMed]
  • Sakai A, Cox MM. RecFOR and RecOR as Distinct RecA Loading Pathways. J Biol Chem. 2009;284:3264–3272. [PMC free article] [PubMed]
  • Slilaty SN, Little JW. Lysine-156 and serine-119 are required for LexA repressor cleavage: a possible mechanism. Proc Natl Acad Sci USA. 1987;84:3987–3991. [PubMed]
  • Sommer S, Knezevic J, Bailone A, Devoret R. Induction of only one SOS operon, umuDC, is required for SOS mutagenesis in Escherichia coli. Mol Gen Genet. 1993;239:137–144. [PubMed]
  • Stohl EA, Brockman JP, Burkle KL, Morimatsu K, Kowalczykowski SC, Siefert HS. Escherichia coli RecX inhibits RecA recombinase and coprotease activities in vitro and in vivo. J Biol Chem. 2003;278:2278–2285. [PubMed]
  • Tanaka K, Hiramoto T, Fukuda T, Miyagawa K. A novel human rad54 homologue, Rad54B, associates with Rad51. J Biol Chem. 2000;275:26316–26321. [PubMed]
  • Veaute X, Delmas P, Selva M, Jeusset J, Le Cam E, Matic I, Fabre F, Petit MA. UvrD helicase, unlike Rep helicase, dismantles RecA nucleoprotein filaments in Escherichia coli. EMBO J. 2005;24:180–189. [PubMed]
  • Walker GC, Smith BT, Sutton MD. The SOS response to DNA damage. In: Storz G, HenggeAronis R, editors. Bacterial Stress Responses. Washington, D.C.: American Society of Microbiology; 2000. pp. 131–144.
  • Wolkow CA, DeBoy RT, Craig NL. Conjugating plasmids are preferred targets for Tn7. Genes Dev. 1996;10:2145–2157. [PubMed]
  • Yuan ZM, Huang Y, Ishiko T, Nakada S, Utsugisawa T, Kharbanda S, Wang R, Sung P, Shinohara ARW, Kufe D. Regulation of Rad51 function by c-Abl in response to DNA damage. J Biol Chem. 1998;273:3799–3802. [PubMed]