Rep Provides a Second Motor at the Replisome to Promote Duplication of Protein-Bound DNA

Summary Nucleoprotein complexes present challenges to genome stability by acting as potent blocks to replication. One attractive model of how such conflicts are resolved is direct targeting of blocked forks by helicases with the ability to displace the blocking protein-DNA complex. We show that Rep and UvrD each promote movement of E. coli replisomes blocked by nucleoprotein complexes in vitro, that such an activity is required to clear protein blocks (primarily transcription complexes) in vivo, and that a polarity of translocation opposite that of the replicative helicase is critical for this activity. However, these two helicases are not equivalent. Rep but not UvrD interacts physically and functionally with the replicative helicase. In contrast, UvrD likely provides a general means of protein-DNA complex turnover during replication, repair, and recombination. Rep and UvrD therefore provide two contrasting solutions as to how organisms may promote replication of protein-bound DNA.

bacteriophage T4 DNA (generously provided by Richard Bowater, University of East Anglia) with primers encoding NcoI and XbaI sites at the 5' and 3' ends of the gene, respectively, and cloned into pBAD cut with NcoI and XbaI. DNA sequencing revealed no errors within the cloned dda. pBADrecD2 was generated by amplification of Deinococcus radiodurans recD2 from a pET22b clone in which codon usage had been optimised for expression in E. coli (Saikrishnan et al., 2008). Amplification was performed with primers encoding NcoI and SmaI sites, the PCR product was cloned into pBAD cleaved with the same two restriction enzymes and the clone checked by DNA sequencing.
A rep overexpression clone encoding an N-terminal peptide tag for biotinylation by BirA was constructed by PCR amplification of rep from MG1655 and cloning into pET22b as a NdeI-XhoI fragment. Complementary oligonucleotides, encoding the biotinylation target MSGLNDIFEAQKIEWHEGGG, were then annealed and cloned into the NdeI site of pET22brep to generate pET22bbiorep. A repΔ2B expression clone was generated by PCR amplification of repΔ2B using pRepOΔ2B as template (Cheng et al., 2002) and primers bearing BamHI and NdeI sites at the 5' end and XhoI and PstI sites at the 3' end. This amplification product was cloned as a BamHI-PstI fragment into pUC19 cleaved with the same enzymes to yield pPM676 from which the EcoRI-XhoI rep fragment was cloned into pET22bbiorep cleaved with EcoRI and XhoI to generate pET22bbiorepΔ2B. To generate a rep2B expression clone, the rep sequence corresponding to G373-G543 was amplified using primers incorporating a start codon and NdeI site within the upstream primer and a stop codon and XhoI site within the downstream primer and cloned into pET22b to yield pET22brep2B. Introduction of the biotin target was performed as described above for wild type rep. A plasmid to express biotinylated RepΔC33 was generated by use of oligonucleotides to amplify rep lacking the final 33 codons and bearing BamHI and NdeI sites at the 5' end and XhoI and PstI sites at the 3' end, using pRH72 (Heller and Marians, 2005) as a template. This PCR product was cloned as a BamHI-PstI fragment into pACT2 cleaved with BamHI and XhoI to yield pMG19 from which the EcoRI-XhoI rep fragment was cloned into pET22bbiorep cleaved with the same enzymes to generate pET22bbiorepΔC33. 1995; Payne et al., 2006). TRCF, B. stearothermophilus DnaB and PcrA, bacteriophage T4 Dda and Deinococcus radiodurans RecD2 were kind gifts of Nigel Savery (University of Bristol), Panos Soultanas (University of Nottingham), Kevin Raney (University of Arkansas for Medical Sciences) and Dale Wigley (CRUK). N-terminally biotinylated C33 Rep , equivalent to the C-terminal 33 residues of Rep, was synthesised by Alta Biosciences, Birmingham, UK. EcoRI K62E (EcoRI HF) and EagI were purchased from New England Biolabs.
The nine individual subunits of DNA polymerase III were overexpressed and purified (at 4°C) as follows. The α and ε subunit genes were cloned by PCR amplification into pET22b and pET24a, respectively, and co-overexpressed in BL21(DE3) in the presence of both ampicillin and kanamycin. Cells (8 L) were grown in F medium (Kim and McHenry, 1996) at 37°C to an A 650 of 0.3, the temperature shifted to 25°C and, once the A 650 had reached 0.5, IPTG was added to 1 mM and growth continued for a further 3 1/2 hours. Cells were harvested by centrifugation and resuspended in 50 mM Tris-HCl pH 7.5 and 10% sucrose to a final volume of 60 ml prior to storage at -80°C. Cells were subsequently thawed and then lysed by the addition of EDTA to 20 mM, Tris-HCl pH 8.4 to 50 mM, KCl to 150 mM, DTT to 10 mM, PMSF to 0.1 mM, 215 mg of protease inhibitor cocktail (Sigma) and lysozyme to 0.2 mg/ml. Cells were left on ice for 10 minutes then Brij 58 added to a final concentration of 0.1%. Incubation was continued on ice for a further 30 minutes followed by centrifugation at 100,000g at 4°C for one hour. The supernatant was recovered and Polymin P added dropwise to 0.075% at 4°C with stirring for one hour.
Nucleic acids were precipitated by centrifugation in a Sorvall SS34 rotor at 20,000 rpm for 20 minutes at 4°C. Solid ammonium sulphate was then added to the recovered supernatant to 25% saturation, stirred on ice for 20 minutes and then centrifuged as for the Polymin P precipitation. Ammonium sulphate was then added to 50% saturation and proteins precipitated as above. This pellet was resuspended in 50 mM Tris-HCl pH 7.5, 1 mM EDTA, 1 mM DTT (buffer A) and 0.1 mM PMSF to a final volume of 110 ml. The αε complex was purified using Q-Sepharose, Superdex 200, heparin-agarose, dsDNAcellulose and Mono-Q columns using buffer A plus NaCl. Purified αε was stored in 100 mM NaCl, 50 mM Tris-HCl pH 7.5, 1 mM EDTA, 1 mM DTT and 20% glycerol at -80°C.
The θ subunit gene, holE, was cloned by PCR amplification into pET22b and overexpressed in BL21(DE3). Cells (8 L) were grown at 37°C in LB medium with ampicillin until the A 650 reached 0.4. IPTG was then added to 1 mM and growth continued at 37°C for a further three hours prior to harvesting the cells by centrifugation and resuspension in Tris and sucrose as described above. Lysis and nucleic acid precipitation was as described above, except that the final concentration of Polymin P was 0.4%. The supernatant from the Polymin P precipitation was applied to a Q-Sepharose column equilibrated in buffer B (20 mM Tris-HCl pH 7.5, 20% glycerol, 0.5 mM EDTA, 2 mM DTT) plus 50 mM NaCl. θ eluted in the flow-through and was then applied to a heparin-agarose column equilibrated in buffer B plus 50 mM NaCl. Again θ eluted in the flow-through and was applied to a S-Sepharose column equilibrated in buffer B plus 50 mM NaCl. θ was again present in the flow-through and was then loaded on to a phosphocellulose column equilibrated in buffer C (10 mM sodium phosphate pH 6.5, 10% glycerol, 0.5 mM EDTA, 2 mM DTT). θ was eluted with a linear gradient of 10-200 mM sodium phosphate pH 6.5 in buffer C (Studwell-Vaughan and O'Donnell, 1993). Peak fractions (approximately 100 mM sodium phosphate) were collected, ammonium sulphate added to 50% saturation and precipitated proteins harvested as above. The precipitate was resuspended in buffer B and passed through a Sephacryl S-200 gel filtration column equilibrated in buffer B plus 200 mM NaCl. Peak fractions were pooled and buffer B plus 80% glycerol was added to give a final glycerol concentration of 50% prior to storage at -80°C.
The γ and τ subunits are expressed from the same gene, dnaX, with γ being a truncated form of τ (Blinkova et al., 1993;Kodaira et al., 1983). γ was expressed by cloning the γ-only reading frame into pET22b. Overexpression was performed in BL21(DE3)/pLysS by growth of cells in 8 L of F medium with ampicillin and chloramphenicol at 37°C until the A 650 reached 0.6. IPTG was added to 1 mM and incubation continued at 37°C for a further three hours. Cells were harvested and resuspended in Tris and sucrose as above. Cells were lysed and treated with 0.075% Polymin P as described above, then proteins within the supernatant were precipitated by addition of ammonium sulphate to 40%. The precipitate was backwashed with 50 mM Tris-HCl pH 7.5, 1 mM EDTA, 10% glycerol, 5 mM DTT (buffer D) plus 50 mM NaCl and 0.1 mM PMSF with 35% then 30% ammonium sulphate. The final pellet was resuspended in buffer D and purified through Q-Sepharose, Sephacryl S200, heparin-agarose and DEAE-Sepharose in buffer D plus NaCl. γ bound to all but heparin-agarose. Fractions from the DEAE-Sepharose column were pooled and dialysed into buffer D plus 200 mM NaCl and 40% glycerol prior to storage at -80°C.
The τ subunit was overexpressed from the wild type dnaX gene cloned into pET22b.
Overexpression was performed in BL21(DE3). Cells were grown in 8 L of F medium plus ampicillin at 37°C until an A 650 of 0.6 was reached. IPTG was added to 1 mM and growth continued at 37°C for three hours. Cells were harvested and resuspended as above. Cells were lysed and treated with 0.075% Polymin P as described above, then ammonium sulphate was added to 40% saturation. After centrifugation, the pellet was resuspended in buffer A and then backwashed with 35% then 30% ammonium sulphate in buffer A. The final backwashed pellet was resuspended in buffer A plus 50 mM NaCl, 0.1 mM PMSF and 55 mg of protease inhibitor cocktail. τ was purified on S-Sepharose, heparin-agarose and Superdex 200 columns in buffer A plus NaCl. To remove traces of γ, pooled fractions from the gel filtration step were loaded onto a Mono-S HR 5/5 column equilibrated in buffer A plus 50 mM NaCl. τ was eluted with a 20 column volume gradient of 50 to 300 mM NaCl in buffer A and dialysed against 100 mM NaCl, 50 mM Tris-HCl pH 7.5, 1 mM EDTA, 1 mM DTT and 50% glycerol prior to storage at -80°C.
The χ and ψ subunit genes were cloned into pET24a and pET22b, respectively, and co-overexpressed in BL21(DE3) in the presence of ampicillin and kanamycin. Cells (8 L) were grown in LB medium at 37°C to an A 650 of 0.8 at which point IPTG was added to 1 mM and growth continued for a further three hours at 37°C. Cells were harvested and resuspended in Tris and sucrose as described above. Lysis was performed as above and nucleic acids were precipitated with 0.4% Polymin P. Proteins were precipitated by addition of solid ammonium sulphate to the supernatant to 45% saturation. The resultant pellet was resuspended in 50 mM Tris-HCl pH 7.8, 1 mM EDTA, 5 mM DTT and 20% glycerol (buffer E) and dialysed against buffer E plus 20 mM NaCl. The dialysate was loaded onto a Q-Sepharose column equilibrated in buffer E plus 20 mM NaCl and eluted with a 10 column volume linear gradient of 20 to 200 mM NaCl in buffer E. Peak fractions containing χ and ψ were pooled and dialysed in 50 mM Tris-HCl pH 7, 1 mM EDTA, 5 mM DTT and 20 mM NaCl (buffer F) prior to loading onto a S-Sepharose column equilibrated in buffer F plus 20 mM NaCl. Proteins were eluted with a 10 column volume gradient of 20 to 200 mM NaCl in buffer F. Peak fractions containing χ and ψ were pooled and proteins were precipitated by addition of solid ammonium sulphate to 45% saturation. The pellet was resuspended in buffer F and then passed through a Sephacryl S-200 gel filtration column equilibrated in buffer F plus 20 mM NaCl. Peak fractions were pooled and dialysed against buffer F plus 20 mM NaCl and 50% glycerol prior to storage at -80°C.
The δ subunit gene was cloned into pET22b and overexpressed in BL21(DE3). Cells (8 L) were grown in LB with ampicillin at 37°C to an A 650 of 0.5. IPTG was added to 1 mM and growth continued for a further three hours at 37°C before harvesting and storage of cells as described above. Cells were then thawed, lysed, and nucleic acids precipitated with 0.4% Polymin P as described above. Proteins were precipitated with 50% ammonium sulphate and resuspended in 50 mM Tris-HCl pH 8, 1 mM EDTA and 20% glycerol (buffer G) to an equivalent conductivity of buffer G plus 50 mM NaCl. δ was then purified on Q-Sepharose, heparin-agarose and Sephacryl S-200 columns in buffer G plus NaCl prior to storage in 50 mM Tris-HCl pH 7.5, 75 mM NaCl, 1 mM EDTA, 5 mM DTT and 50% glycerol.
The δ' subunit gene was cloned into pET22b and overexpressed as described for δ.
Lysis, Polymin P and ammonium sulphate precipitation was also as for δ. Protein pellets after ammonium sulphate precipitation were resuspended in 50 mM Tris-HCl pH 8.5, 1 mM EDTA and 5 mM DTT (buffer H) and then dialysed against buffer H. δ' was purified by chromatography through Q-Sepharose, Sephacryl S-200 and heparin-agarose in buffer H plus NaCl prior to storage in 50 mM Tris-HCl pH 8, 50 mM NaCl, 1 mM EDTA, 2.5 mM DTT and 50% glycerol.
Biotinylated Rep was overexpressed by transformation of pET22bbiorep into E. coli BL21(DE3) harbouring pBirAcm (Avidity, LLC), encoding a biotin ligase. Colonies containing both the pET22b and pBirAcm plasmids were selected using ampicillin (100 μg ml -1 ) and chloramphenicol (25 μg ml -1 ). A single colony was inoculated into 10 ml of LB broth containing ampicillin and chloramphenicol and grown for 8 hours at 37°C with shaking. Cells were harvested by centrifugation, resuspended in 10 ml of LB and 1 ml inoculated into 80 ml of LB containing ampicillin and chloramphenicol. After growth at 37°C with shaking overnight, the cells were pelleted, resuspended in fresh LB and inoculated into 8 l of LB in a stirred vessel fermenter containing the above antibiotics and grown at 37°C to an A 650 of 0.4. The culture was allowed to cool to 25°C and then IPTG and biotin were added to final concentrations of 1 mM and 500 μM, respectively, and incubation continued at 25°C for a further 3 hours. Cells were harvested by centrifugation and the cell pellet resuspended in 50 mM Tris-HCl pH 7.5 and 10% sucrose prior to storage at -80°C.
Cells equivalent to 2 l of the culture were subsequently thawed and then lysed by the addition of EDTA to 1 mM, Tris-HCl pH 8.4 to 50 mM, KCl to 150 mM, DTT to 1 mM, PMSF to 1 mM and lysozyme to 0.2 mg/ml. Cells were left on ice for 10 minutes then Brij 58 added to a final concentration of 0.1%. Incubation was continued on ice for a further 30 minutes followed by centrifugation at 100,000g at 4°C for one hour. The supernatant was recovered and solid ammonium sulphate was then added to the recovered supernatant to 50% saturation, stirred on ice for 20 minutes and then centrifuged in a Sorvall SS34 rotor at 20,000 rpm for 20 minutes at 4°C. Precipitated protein was resuspended in buffer A (see above) plus 0.1 mM PMSF until a conductance equivalent to buffer A plus 100 mM NaCl was reached. The resuspension was loaded onto a 5 ml Softlink Avidin column (Promega) equilibrated in buffer A plus 100 mM NaCl. The column was washed first with buffer A plus 100 mM NaCl, then buffer A plus 1 M NaCl and finally buffer A plus 100 mM NaCl. Biotinylated protein was then eluted by flushing the column with buffer A containing 100 mM NaCl and 5 mM biotin, pausing the elution for 30 minutes just as protein began to appear in the eluate and then continuing the elution with the same buffer. Fractions containing biotinylated Rep were pooled and loaded onto a 1 ml Hi-trap heparin column (GE Healthcare) equilibrated in buffer A plus 100 mM NaCl, the column was washed with buffer A plus 100 mM NaCl and then protein eluted using a 100-1000 mM NaCl gradient.
Biotinylated RepΔ2B, Rep2B, RepΔC33 and UvrD were overexpressed and purified in a similar manner except that all 8 l of the RepΔ2B culture was used for purification.
RepΔ2B was also purified in a similar manner except that chromatography on heparin was replaced by Mono-Q (GE Healthcare).

Reconstitution of Core Polymerase and Clamp Loader Complexes
Core polymerase was reconstituted by mixing 1 mg of αε with 0.4 mg of θ in 20 mM Tris-HCl pH 7.5, 2 mM DTT, 0.5 mM EDTA and 20% glycerol (buffer I) plus 30 mM NaCl.
Proteins were incubated at 4°C with gentle agitation for one hour. The mixture was then loaded onto a Mono-Q HR 5/5 column equilibrated in buffer I and αεθ eluted with a 40 column volume linear gradient of 0-400 mM NaCl (Onrust et al., 1995) in buffer I. Fractions were pooled and stored at -80°C in 20 mM Tris-HCl pH 7.5, 1 mM EDTA, 2 mM DTT, 180 mM NaCl and 40% glycerol.
τ clamp loader was reconstituted by mixing 1.1 mg of τ, 0.88 mg of δ, 0.87 mg of δ', and 1.1 mg of χψ in buffer I and incubating at 4°C with gentle agitation for one hour. The mixture was then purified on a Mono-Q HR 5/5 column as for core polymerase except that 80 column volumes were used for elution. Fractions containing all five proteins were pooled and stored in 50 mM Tris-HCl pH 7.5, 1 mM DTT, 1 mM EDTA, 100 mM NaCl and 50% glycerol at -80°C. γ clamp loader was reconstituted as for the τ complex except that the τ subunit was replaced by 1.55 mg of γ.

EcoRI Cleavage Assay
To establish whether Rep or UvrD could displace E111G from unreplicated DNA, µl. Reactions were preincubated for 2 minutes at 37°C prior to addition of EcoRI E111G, Rep and/or UvrD. Where E111G was added together with Rep or UvrD, proteins were premixed prior to addition to the reaction. Incubation was continued for 1 minute at 37°C.
EcoRI K62E ("EcoRI-HF" from New England Biolabs Inc.) without and with E111G was then added and incubation continued for 1 minute at 37°C. In the reaction containing both restriction enzymes ( Figure S1B, lane 9), the proteins were premixed prior to addition. EDTA, 150 mM NaCl, 10 mM MgCl 2 and 0.005% Tween 20 at 20 μl min -1 . This buffer differed from that used in the in vitro DNA replication assays so as to minimise nonspecific interactions with the surface-immobilised streptavidin.

Pulldown Assays
Cell lysate was prepared by growth of E. coli MG1655 in 1 l of LB at 37°C with shaking to an A 600 of 1.2, harvesting of cells by centrifugation and resuspension in 20 ml of 10% sucrose, 20 mM Tris-HCl pH 7.5, 1 mM EDTA, 1 mM DTT and 100 mM NaCl. Cells were lysed by sonication followed by addition of lysozyme to 0.5 mg ml -1 and Triton X-100 to 0.1% and subsequent agitation at room temperature for 20 minutes. The lysate was then clarified by centrifugation.
Pulldown experiments were performed using streptavidin-coated magnetic beads Western blots were performed on gels in which 5 rather than 30 μl of supernatant was electrophoresed using rabbit antisera raised against purified DnaB together with HRPconjugated mouse anti-rabbit antibodies and ECLPlus reagents (GE Healthcare).
Coomassie-stained gels also contained 100 ng purified DnaB whilst gels for Western blotting contained 40 ng purified DnaB.

Bandshift Assays
A radioactively labelled forked DNA substrate having a 60 bp duplex and two 38 base ssDNA arms was formed from oligonucleotides 5 and 6 as described in (Atkinson et al., 2009). Bandshifts were performed in 50 mM HEPES pH 8, 10 mM magnesium acetate, 10 mM DTT, 10 μM ADP and 50 μg ml -1 bovine serum albumin. Reactions containing 1 nM DNA substrate were preincubated for 2 minutes at 37°C prior to addition of proteins at the indicated concentrations to give a final volume of 10 μl. Incubation was continued for 10 minutes at 37°C prior to addition of 2 μl of 30% glycerol. Reactions were then loaded onto a 4% polyacrylamide gel with 89 mM Tris base, 89 mM boric acid and 10 μM ADP as running buffer this and then electrophoresis performed at 160 V for 90 minutes at room temperature prior to drying and analysis by autoradiography and phosphorimaging.

Fork Unwinding Assays
The forked DNA substrate to analyse cooperativity between Rep and DnaB was that used for the bandshift assays described above. performed in a similar manner except that the forked DNA substrate was formed from oligonucleotides 1 and 2 described in (Cadman et al., 2006).

Figure S1. Blockage of Replication Forks In Vitro by EcoRI E111G-DNA Complexes
(A) Denaturing agarose gel of replication products formed with oriC-containing plasmids bearing 0, 2 or 8 EcoRI sites performed as shown in Figure 1A except that reactions were terminated prior to addition of unlabelled dCTP and candidate helicases. Reactions contained 0, 25, 100 or 200 nM E111G dimers, as indicated. Sizes of HindIII-cut λDNA, in kb, are noted on the left and estimated sizes of leading strand products on the right. Note that whilst the full length leading strands generated from the plasmids bearing 2 and 8 EcoRI sites (pME101 and pPM594 respectively) were 4.7 kb, the full length leading strand generated from pPM436 containing no EcoRI sites was 4.4 kb.  (A) pME101, containing 2 EcoRI sites and as used in Figure 1D, was linearised by cleavage with EagI as for the in vitro replication reactions and processed as indicated.
(B) Ethidium bromide-stained agarose gel of EcoRI K62E cleavage products obtained from the linearised duplex. The purified linearised duplex was incubated with E111G, Rep and/or UvrD for 1 minute at 37°C. EcoRI K62E with and without E111G was then added and incubation continued at 37°C for a further 1 minute prior to deproteinisation of the reactions and electrophoresis. EcoRI K62E (New England Biolabs Inc.) has the same cleavage specificity as wild type EcoRI but has reduced star activity, with cleavage of the linear duplexes generating the expected 4.5 and 1.2 kb products (lane 3). Note that the 0.3 kb product was not resolved under these conditions. No cleavage of the template was observed with E111G (compare lanes 1 and 2), as expected (King et al., 1989).
Preincubation of the DNA with E111G prior to addition of EcoRI K62E resulted in significant inhibition of cleavage (compare lanes 3 and 4) demonstrating that binding of E111G to the EcoRI sites inhibited access by EcoRI K62E. Addition of Rep or UvrD simultaneously with E111G provided no detectable relief of this inhibition (lanes 5 and 6).
This lack of relief was not due to direct inhibition of EcoRI K62E cleavage by Rep or UvrD as neither helicase inhibited cleavage in the absence of E111G (lanes 7 and 8).
Furthermore, high levels of cleavage were still obtained upon simultaneous addition of E111G and EcoRI K62E, although cleavage was reduced compared to those obtained in the absence of E111G (compare lanes 3 and 9). EcoRI K62E could therefore compete with E111G in the absence of prior binding of E111G. The inability of either Rep or UvrD to promote cleavage indicates therefore that neither helicase can promote efficient displacement of E111G from unreplicated DNA.

Figure S3.
Interaction of surface immobilised biotinylated Rep, a mutant Rep protein lacking the 2B subdomain (RepΔ2B) and the isolated Rep 2B subdomain (Rep2B) with E. coli DnaB.
8100, 7600 and 7100 resonance units of Rep, RepΔ2B and Rep2B were surface immobilised onto streptavidin-coated chips. DnaB was present at 4 μM monomers.The Rep2B subdomain provides an autoregulatory mechanism to control DNA unwinding, possibly mediated via unidentified protein-protein interactions (Brendza et al., 2005).
However, RepΔ2B retained the ability to interact with DnaB whilst there was no detectable interaction between the isolated 2B subdomain and DnaB. Interaction between DnaB and Rep was not therefore mediated via the Rep 2B subdomain.      (Bachmann, 1996) CGSC4518 pyrE60 (Cherepanov and Wackernagel, 1995) (Bernhardt and de Boer, 2004) a Only the relevant additional genotype of the derivatives is shown b A new deletion allele of uvrD (uvrD::dhfr) was made using one-step gene inactivation (Datsenko and Wanner, 2000). The entire coding sequence was deleted and replaced with a sequence encoding resistance to trimethoprim.