Gene doctoring: a method for recombineering in laboratory and pathogenic Escherichia coli strains

Background  

Homologous recombination mediated by the λ-Red genes is a common method for making chromosomal modifications in Escherichia coli. Several protocols have been developed that differ in the mechanisms by which DNA, carrying regions homologous to the chromosome, are delivered into the cell. A common technique is to electroporate linear DNA fragments into cells. Alternatively, DNA fragments are generated in vivo by digestion of a donor plasmid with a nuclease that does not cleave the host genome. In both cases the λ-Red gene products recombine homologous regions carried on the linear DNA fragments with the chromosome. We have successfully used both techniques to generate chromosomal mutations in E. coli K-12 strains. However, we have had limited success with these λ-Red based recombination techniques in pathogenic E. coli strains, which has led us to develop an enhanced protocol for recombineering in such strains.     

Lambda Red Mediated Gap Repair Utilizes a Novel Replicative Intermediate in Escherichia coli

The lambda phage Red recombination system can mediate efficient homologous recombination in Escherichia coli, which is the basis of the DNA engineering technique termed recombineering. Red mediated insertion of DNA requires DNA replication, involves a single-stranded DNA intermediate and is more efficient on the lagging strand of the replication fork. Lagging strand recombination has also been postulated to explain the Red mediated repair of gapped plasmids by an Okazaki fragment gap filling model. Here, we demonstrate that gap repair involves a different strand independent mechanism. Gap repair assays examining the strand asymmetry of recombination did not show a lagging strand bias. Directly testing an ssDNA plasmid showed lagging strand recombination is possible but dsDNA plasmids did not employ this mechanism. Insertional recombination combined with gap repair also did not demonstrate preferential lagging strand bias, supporting a different gap repair mechanism. The predominant recombination route involved concerted insertion and subcloning though other routes also operated at lower frequencies. Simultaneous insertion of DNA resulted in modification of both strands and was unaffected by mutations to DNA polymerase I, responsible for Okazaki fragment maturation. The lower efficiency of an alternate Red mediated ends-in recombination pathway and the apparent lack of a Holliday junction intermediate suggested that gap repair does not involve a different Red recombination pathway. Our results may be explained by a novel replicative intermediate in gap repair that does not involve a replication fork. We exploited these observations by developing a new recombineering application based on concerted insertion and gap repair, termed SPI (subcloning plus insertion). SPI selected against empty vector background and selected for correct gap repair recombinants. We used SPI to simultaneously insert up to four different gene cassettes in a single recombineering reaction. Consequently, our findings have important implications for the understanding of E. coli replication and Red recombination.     

Subcloning Plus Insertion (SPI) - A Novel Recombineering Method for the Rapid Construction of Gene Targeting Vectors

Gene targeting refers to the precise modification of a genetic locus using homologous recombination. The generation of novel cell lines and transgenic mouse models using this method necessitates the construction of a ‘targeting’ vector, which contains homologous DNA sequences to the target gene, and has for many years been a limiting step in the process. Vector construction can be performed in vivo in Escherichia coli cells using homologous recombination mediated by phage recombinases using a technique termed recombineering. Recombineering is the preferred technique to subclone the long homology sequences (>4kb) and various targeting elements including selection markers that are required to mediate efficient allelic exchange between a targeting vector and its cognate genomic locus. Typical recombineering protocols follow an iterative scheme of step-wise integration of the targeting elements and require intermediate purification and transformation steps. Here, we present a novel recombineering methodology of vector assembly using a multiplex approach. Plasmid gap repair is performed by the simultaneous capture of genomic sequence from mouse Bacterial Artificial Chromosome libraries and the insertion of dual bacterial and mammalian selection markers. This subcloning plus insertion method is highly efficient and yields a majority of correct recombinants. We present data for the construction of different types of conditional gene knockout, or knock-in, vectors and BAC reporter vectors that have been constructed using this method. SPI vector construction greatly extends the repertoire of the recombineering toolbox and provides a simple, rapid and cost-effective method of constructing these highly complex vectors.     

Probing cellular processes with oligo-mediated recombination and using the knowledge gained to optimize recombineering

Recombination with single-strand DNA oligonucleotides (oligos) in Escherichia coli is an efficient and rapid way to modify replicons in vivo. The generation of nucleotide alteration by oligo recombination provides novel assays for studying cellular processes. Single-strand exonucleases inhibit oligo recombination, and recombination is increased by mutating all four known exonucleases. Increasing oligo concentration or adding nonspecific carrier oligo titrates out the exonucleases. In a model for oligo recombination, λ Beta protein anneals the oligo to complementary single-strand DNA at the replication fork. Mismatches are created, and the methyl-directed mismatch repair (MMR) system acts to eliminate the mismatches inhibiting recombination. Three ways to evade MMR through oligo design include, in addition to the desired change (1) a C·C mismatch  6 bp from that change; (2) four or more adjacent mismatches; or (3) mismatches at four or more consecutive wobble positions. The latter proves useful for making high-frequency changes that alter only the target amino acid sequence and even allows modification of essential genes. Efficient uptake of DNA is important for oligo-mediated recombination. Uptake of oligos or plasmids is dependent on media and is 10,000-fold reduced for cells grown in minimal versus rich medium. Genomewide engineering technologies utilizing recombineering will benefit from both optimized recombination frequencies and a greater understanding of how biological processes such as DNA replication and cell division impact recombinants formed at multiple chromosomal loci. Recombination events at multiple loci in individual cells are described here.     

Cas9-assisted recombineering in C. elegans: genome editing using in vivo assembly of linear DNAs

Recombineering, the use of endogenous homologous recombination systems to recombine DNA in vivo, is a commonly used technique for genome editing in microbes. Recombineering has not yet been developed for animals, where non-homology-based mechanisms have been thought to dominate DNA repair. Here, we demonstrate, using Caenorhabditis elegans, that linear DNAs with short homologies (∼35 bases) engage in a highly efficient gene conversion mechanism. Linear DNA repair templates with homology to only one side of a double-strand break (DSB) initiate repair efficiently, and short overlaps between templates support template switching. We demonstrate the use of single-stranded, bridging oligonucleotides (ssODNs) to target PCR fragments for repair of DSBs induced by CRISPR/Cas9 on chromosomes. Based on these findings, we develop recombineering strategies for precise genome editing that expand the utility of ssODNs and eliminate in vitro cloning steps for template construction. We apply these methods to the generation of GFP knock-in alleles and gene replacements without co-integrated markers. We conclude that, like microbes, metazoans possess robust homology-dependent repair mechanisms that can be harnessed for recombineering and genome editing.     

Recombineering with tolC as a selectable/counter-selectable marker: remodeling the rRNA operons of Escherichia coli

This work describes the novel use of tolC as a selectable/counter-selectable marker for the facile modification of DNA in Escherichia coli. Expression of TolC (an outer membrane protein) confers relative resistance to toxic small molecules, while its absence renders the cell tolerant to colicin E1. These features, coupled with the λredgam recombination system, allow for selection of tolC insertions/deletions anywhere on the E. coli chromosome or on plasmid DNA. This methodology obviates the need for minimal growth media, specialized wash protocols and the lengthy incubation times required by other published recombineering methods. As a rigorous test of the TolC selection system, six out of seven 23S rRNA genes were consecutively and seamlessly removed from the E. coli chromosome without affecting expression of neighboring genes within the complex rrn operons. The resulting plasmid-free strain retains one 23S rRNA gene (rrlC) in its natural location on the chromosome and is the first mutant of its kind. These new rRNA mutants will be useful in the study of rRNA gene regulation and ribosome function. Given its high efficiency, low background and facility in rich media, tolC selection is a broadly applicable method for the modification of DNA by recombineering.     

Coupling the CRISPR/Cas9 System with Lambda Red Recombineering Enables Simplified Chromosomal Gene Replacement in Escherichia coli

To date, most genetic engineering approaches coupling the type II Streptococcus pyogenes clustered regularly interspaced short palindromic repeat (CRISPR)/Cas9 system to lambda Red recombineering have involved minor single nucleotide mutations. Here we show that procedures for carrying out more complex chromosomal gene replacements in Escherichia coli can be substantially enhanced through implementation of CRISPR/Cas9 genome editing. We developed a three-plasmid approach that allows not only highly efficient recombination of short single-stranded oligonucleotides but also replacement of multigene chromosomal stretches of DNA with large PCR products. By systematically challenging the proposed system with respect to the magnitude of chromosomal deletion and size of DNA insertion, we demonstrated DNA deletions of up to 19.4 kb, encompassing 19 nonessential chromosomal genes, and insertion of up to 3 kb of heterologous DNA with recombination efficiencies permitting mutant detection by colony PCR screening. Since CRISPR/Cas9-coupled recombineering does not rely on the use of chromosome-encoded antibiotic resistance, or flippase recombination for antibiotic marker recycling, our approach is simpler, less labor-intensive, and allows efficient production of gene replacement mutants that are both markerless and “scar”-less.     

Precision genome engineering in lactic acid bacteria

Innovative new genome engineering technologies for manipulating chromosomes have appeared in the last decade. One of these technologies, recombination mediated genetic engineering (recombineering) allows for precision DNA engineering of chromosomes and plasmids in Escherichia coli. Single-stranded DNA recombineering (SSDR) allows for the generation of subtle mutations without the need for selection and without leaving behind any foreign DNA. In this review we discuss the application of SSDR technology in lactic acid bacteria, with an emphasis on key factors that were critical to move this technology from E. coli into Lactobacillus reuteri and Lactococcus lactis. We also provide a blueprint for how to proceed if one is attempting to establish SSDR technology in a lactic acid bacterium. The emergence of CRISPR-Cas technology in genome engineering and its potential application to enhancing SSDR in lactic acid bacteria is discussed. The ability to perform precision genome engineering in medically and industrially important lactic acid bacteria will allow for the genetic improvement of strains without compromising safety.     

FRUIT,a Scar-Free System for Targeted Chromosomal Mutagenesis,Epitope Tagging,and Promoter Replacement in Escherichia coli and Salmonella enterica

Recombineering is a widely-used approach to delete genes, introduce insertions and point mutations, and introduce epitope tags into bacterial chromosomes. Many recombineering methods have been described, for a wide range of bacterial species. These methods are often limited by (i) low efficiency, and/or (ii) introduction of “scar” DNA into the chromosome. Here, we describe a rapid, efficient, PCR-based recombineering method, FRUIT, that can be used to introduce scar-free point mutations, deletions, epitope tags, and promoters into the genomes of enteric bacteria. The efficiency of FRUIT is far higher than that of the most widely-used recombineering method for Escherichia coli. We have used FRUIT to introduce point mutations and epitope tags into the chromosomes of E. coli K-12, Enterotoxigenic E. coli, and Salmonella enterica. We have also used FRUIT to introduce constitutive and inducible promoters into the chromosome of E. coli K-12. Thus, FRUIT is a versatile, efficient recombineering approach that can be applied in multiple species of enteric bacteria.     

Lambda Red-mediated recombinogenic engineering of enterohemorrhagic and enteropathogenic <Emphasis Type="Italic">E. coli</Emphasis>

Background

The λ Red recombineering technology has been used extensively in Escherichia coli and Salmonella typhimurium for easy PCR-mediated generation of deletion mutants, but less so in pathogenic species of E. coli such as EHEC and EPEC. Our early experiments with the use of λ Red in EHEC and EPEC have led to sporadic results, leading to the present study to identify factors that might improve the efficiency of Red recombineering in these pathogenic strains of E. coli.

Results

In this report, we have identified conditions that optimize the use of λ Red for recombineering in EHEC and EPEC. Using plasmids that contain a P_tac-red-gam operon and a temperature-sensitive origin of replication, we have generated multiple mutations (both marked and unmarked) in known virulence genes. In addition, we have easily deleted five O157-specific islands (O-islands) of EHEC suspected of containing virulence factors. We have examined the use of both PCR-generated substrates (40 bp of flanking homology) and plasmid-derived substrates (~1 kb of flanking homology); both work well and each have their own advantages. The establishment of the hyper-rec phenotype requires only a 20 minute IPTG induction period of red and gam. This recombinogenic window is important as constitutive expression of red and gam induces a 10-fold increase in spontaneous resistance to rifampicin. Other factors such as the orientation of the drug marker in recombination substrates and heat shock effects also play roles in the success of Red-mediated recombination in EHEC and EPEC.

Conclusions

The λ Red recombineering technology has been optimized for use in pathogenic species of E. coli, namely EHEC and EPEC. As demonstration of this technology, five O-islands of EHEC were easily and precisely deleted from the chromosome by electroporation with PCR-generated substrates containing drug markers flanked with 40 bp of target DNA. These results should encourage the use of λ Red recombineering in these and other strains of pathogenic bacteria for faster identification of virulence factors and the speedy generation of bacterial mutants for vaccine development.

     

Recombineering Homologous Recombination Constructs in Drosophila

The continued development of techniques for fast, large-scale manipulation of endogenous gene loci will broaden the use of Drosophila melanogaster as a genetic model organism for human-disease related research. Recent years have seen technical advancements like homologous recombination and recombineering. However, generating unequivocal null mutations or tagging endogenous proteins remains a substantial effort for most genes. Here, we describe and demonstrate techniques for using recombineering-based cloning methods to generate vectors that can be used to target and manipulate endogenous loci in vivo. Specifically, we have established a combination of three technologies: (1) BAC transgenesis/recombineering, (2) ends-out homologous recombination and (3) Gateway technology to provide a robust, efficient and flexible method for manipulating endogenous genomic loci. In this protocol, we provide step-by-step details about how to (1) design individual vectors, (2) how to clone large fragments of genomic DNA into the homologous recombination vector using gap repair, and (3) how to replace or tag genes of interest within these vectors using a second round of recombineering. Finally, we will also provide a protocol for how to mobilize these cassettes in vivo to generate a knockout, or a tagged gene via knock-in. These methods can easily be adopted for multiple targets in parallel and provide a means for manipulating the Drosophila genome in a timely and efficient manner.     

Improved seamless mutagenesis by recombineering using ccdB for counterselection

Recombineering, which is the use of homologous recombination for DNA engineering in Escherichia coli, usually uses antibiotic selection to identify the intended recombinant. When combined in a second step with counterselection using a small molecule toxin, seamless products can be obtained. Here, we report the advantages of a genetic strategy using CcdB as the counterselectable agent. Expression of CcdB is toxic to E. coli in the absence of the CcdA antidote so counterselection is initiated by the removal of CcdA expression. CcdB counterselection is robust and does not require titrations or experiment-to-experiment optimization. Because counterselection strategies necessarily differ according to the copy number of the target, we describe two variations. For multi-copy targets, we use two E. coli hosts so that counterselection is exerted by the transformation step that is needed to separate the recombined and unrecombined plasmids. For single copy targets, we put the ccdA gene onto the temperature-sensitive pSC101 Red expression plasmid so that counterselection is exerted by the standard temperature shift to remove the expression plasmid. To reduce unwanted intramolecular recombination, we also combined CcdB counterselection with Redα omission. These options improve the use of counterselection in recombineering with BACs, plasmids and the E. coli chromosome.     

Interplay of DNA repair,homologous recombination,and DNA polymerases in resistance to the DNA damaging agent 4-nitroquinoline-1-oxide in Escherichia coli

Escherichia coli has three DNA damage-inducible DNA polymerases: DNA polymerase II (Pol II), DNA polymerase IV (Pol IV), and DNA polymerase V (Pol V). While the in vivo function of Pol V is well understood, the precise roles of Pol IV and Pol II in DNA replication and repair are not as clear. Study of these polymerases has largely focused on their participation in the recovery of failed replication forks, translesion DNA synthesis, and origin-independent DNA replication. However, their roles in other repair and recombination pathways in E. coli have not been extensively examined. This study investigated how E. coli's inducible DNA polymerases and various DNA repair and recombination pathways function together to convey resistance to 4-nitroquinoline-1-oxide (NQO), a DNA damaging agent that produces replication blocking DNA base adducts. The data suggest that full resistance to this compound depends upon an intricate interplay among the activities of the inducible DNA polymerases and recombination. The data also suggest new relationships between the different pathways that process recombination intermediates.     

Positive and negative selection using the tetA-sacB cassette: recombineering and P1 transduction in Escherichia coli

The two-step process of selection and counter-selection is a standard way to enable genetic modification and engineering of bacterial genomes using homologous recombination methods. The tetA and sacB genes are contained in a DNA cassette and confer a novel dual counter-selection system. Expression of tetA confers bacterial resistance to tetracycline (Tc^R) and also causes sensitivity to the lipophillic chelator fusaric acid; sacB causes sensitivity to sucrose. These two genes are introduced as a joint DNA cassette into Escherichia coli by selection for Tc^R. A medium containing both fusaric acid and sucrose has been developed, in which, coexpression of tetA-sacB is orders of magnitude more sensitive as a counter-selection agent than either gene alone. In conjunction with the homologous recombination methods of recombineering and P1 transduction, this powerful system has been used to select changes in the bacterial genome that cannot be directly detected by other counter-selection systems.     

Recombineering Using RecTE from Pseudomonas syringae

In this report, we describe the identification of functions that promote genomic recombination of linear DNA introduced into Pseudomonas cells by electroporation. The genes encoding these functions were identified in Pseudomonas syringae pv. syringae B728a based on similarity to the lambda Red Exo/Beta and RecET proteins encoded by the lambda and Rac bacteriophages of Escherichia coli. The ability of the pseudomonad-encoded proteins to promote recombination was tested in P. syringae pv. tomato DC3000 using a quantitative assay based on recombination frequency. The results show that the Pseudomonas RecT homolog is sufficient to promote recombination of single-stranded DNA oligonucleotides and that efficient recombination of double-stranded DNA requires the expression of both the RecT and RecE homologs. Additionally, we illustrate the utility of this recombineering system to make targeted gene disruptions in the P. syringae chromosome.There are currently more than 1,500 completed or draft bacterial genome sequences available for public access. This data resource continues to grow rapidly and provides potential insights into the roles of individual genes and regulons. However, testing hypotheses based on sequence data requires direct experimental manipulation of each genome. While many established methods for modifying bacterial DNA can assist in genetic analysis of these organisms, they are often time-consuming and limited with respect to the types of changes that can be directed.New advances in recombineering (genetic engineering by recombination) offer powerful alternative strategies for site-directed mutagenesis of genomic loci and provide methods for rapid and precise functional genomic analysis in some organisms (9, 29, 36-38, 41, 43). In these cases, recombineering is very efficient when phage-encoded recombinases are supplied, such that in vivo expression of these proteins enables direct genetic engineering of chromosomal and episomal replicons. These proteins catalyze RecA-independent recombination (21) of linear DNA substrates with homologous genomic target loci. The phage recombination functions typically involve the coordinated action of a 5′-to-3′ exonuclease (i.e., RecE or lambda Exo) and a single-stranded DNA (ssDNA)-annealing and strand invasion protein (i.e., RecT or lambda Beta), which we shall refer to as recombinases for brevity. The recombinase binds to 3′ ssDNA ends that are exposed by the action of the exonuclease, forming a protein-DNA filament, which protects the substrate DNA and promotes annealing with the homologous genomic sequence (4, 17, 19, 24). The recombinases are sufficient to facilitate recombination of ssDNA oligonucleotides, presumably because the oligonucleotides resemble the 5′-end-resected double-stranded DNA (dsDNA) substrate (11). Most of the recombinase proteins that have been shown to facilitate recombination are located in operons and are adjacent to the exonuclease-encoding genes, although there are cases where functional recombinase proteins have been identified without an accompanying exonuclease (9).Recombineering technologies have great potential in functional genomic applications and have worked exceptionally well in a few species, but adapting current systems to different bacteria is often problematic. Evidence suggests that these recombination systems have narrow species specificity such that a given system may catalyze robust recombination in one species and be essentially nonfunctional when expressed in another (9, 37). The reasons for this are not known but may be due to a requirement for specific interactions between the recombinase and host-encoded factors (9). Although there is a need to apply recombineering techniques to Pseudomonas species, only marginal success using the characterized phage recombination systems has been reported (14, 23). Most notably, recombinant strains of Pseudomonas aeruginosa were generated using long-homology substrates in the presence of plasmids expressing the lambda Red genes, but the relative influence of the Red genes was not reported (23).Here, we describe the identification of new recombineering proteins that function in a pseudomonad. The genes that encode proteins with similarity to the RecE/RecT proteins of the Rac prophage and lambda Red Exo and Beta were identified in Pseudomonas syringae pv. syringae B728a. These proteins promote efficient homologous recombination between genomic loci and linear DNA substrates introduced directly into P. syringae pv. tomato DC3000 cells by electroporation. These findings provide a foundation for more efficient site-directed mutagenesis of chromosomal loci in P. syringae and serve as a strategy for identifying similar proteins for recombineering in other bacteria.     

A Simple and Effective Method for Construction of Escherichia coli Strains Proficient for Genome Engineering

Multiplex genome engineering is a standalone recombineering tool for large-scale programming and accelerated evolution of cells. However, this advanced genome engineering technique has been limited to use in selected bacterial strains. We developed a simple and effective strain-independent method for effective genome engineering in Escherichia coli. The method involves introducing a suicide plasmid carrying the λ Red recombination system into the mutS gene. The suicide plasmid can be excised from the chromosome via selection in the absence of antibiotics, thus allowing transient inactivation of the mismatch repair system during genome engineering. In addition, we developed another suicide plasmid that enables integration of large DNA fragments into the lacZ genomic locus. These features enable this system to be applied in the exploitation of the benefits of genome engineering in synthetic biology, as well as the metabolic engineering of different strains of E. coli.     

A Novel Selection Marker for Efficient DNA Cloning and Recombineering in E. coli

Production of recombinant DNA in bacterial cells is an essential technique in molecular biology. Plasmids are usually maintained in an E. coli host by antibiotic selection. However, there are only a few antibiotic-resistance markers available in common use. Here we report the adoption of a novel selection marker, mfabI (mutant fabI) for plasmid propagation in E. coli. mfabI expands the limited repertoire of selection markers and allows for more efficient molecular manipulation and plasmid propagation in E. coli. We show that mfabI is not only an efficient plasmid selection marker, but it also possesses unique activity that may facilitate molecular manipulation of unstable sequences. Furthermore, we have incorporated mfabI in the recombineering tool kit for generating mouse gene targeting vectors and demonstrate the advantage of using mfabI-containing recombineering vectors.     

An improved recombineering approach by adding RecA to λ Red recombination

Recombineering is the use of homologous recombination in Escherichia coli for DNA engineering. Of several approaches, use of the λ phage Red operon is emerging as the most reliable and flexible. The Red operon includes three components: Redα, a 5′ to 3′ exonuclease, Redβ, an annealing protein, and Redλ, an inhibitor of the major E. coli exonuclease and recombination complex, RecBCD. Most E. coli cloning hosts are recA deficient to eliminate recombination and therefore enhance thestabulity of cloned DNAs. However, loss of RecA also impairs general cellular integrity. Here we report that transient RecA co-expression enhances the total numer of successful recombinations in bacterial artificial chromosomes (BACs), mostly because the E. coli host is more able to survive the stresses of DNA transformation procedures. We combined this practical improvement with the advantages of a temperature-sensitive version of the low copy pSC 101 plasmid to develop a protocol that is convenient and more efficient than any recombineering procedure, for use of either double-or single-stranded DNA, published to date.     

Ultra-Low Background DNA Cloning System

Yeast-based in vivo cloning is useful for cloning DNA fragments into plasmid vectors and is based on the ability of yeast to recombine the DNA fragments by homologous recombination. Although this method is efficient, it produces some by-products. We have developed an “ultra-low background DNA cloning system” on the basis of yeast-based in vivo cloning, by almost completely eliminating the generation of by-products and applying the method to commonly used Escherichia coli vectors, particularly those lacking yeast replication origins and carrying an ampicillin resistance gene (Amp^r). First, we constructed a conversion cassette containing the DNA sequences in the following order: an Amp^r 5′ UTR (untranslated region) and coding region, an autonomous replication sequence and a centromere sequence from yeast, a TRP1 yeast selectable marker, and an Amp^r 3′ UTR. This cassette allowed conversion of the Amp^r-containing vector into the yeast/E. coli shuttle vector through use of the Amp^r sequence by homologous recombination. Furthermore, simultaneous transformation of the desired DNA fragment into yeast allowed cloning of this DNA fragment into the same vector. We rescued the plasmid vectors from all yeast transformants, and by-products containing the E. coli replication origin disappeared. Next, the rescued vectors were transformed into E. coli and the by-products containing the yeast replication origin disappeared. Thus, our method used yeast- and E. coli-specific “origins of replication” to eliminate the generation of by-products. Finally, we successfully cloned the DNA fragment into the vector with almost 100% efficiency.     

Resolution of Joint Molecules by RuvABC and RecG Following Cleavage of the Escherichia coli Chromosome by EcoKI

DNA double-strand breaks can be repaired by homologous recombination involving the formation and resolution of Holliday junctions. In Escherichia coli, the RuvABC resolvasome and the RecG branch-migration enzyme have been proposed to act in alternative pathways for the resolution of Holliday junctions. Here, we have studied the requirements for RuvABC and RecG in DNA double-strand break repair after cleavage of the E. coli chromosome by the EcoKI restriction enzyme. We show an asymmetry in the ability of RuvABC and RecG to deal with joint molecules in vivo. We detect linear DNA products compatible with the cleavage-ligation of Holliday junctions by the RuvABC pathway but not by the RecG pathway. Nevertheless we show that the XerCD-mediated pathway of chromosome dimer resolution is required for survival regardless of whether the RuvABC or the RecG pathway is active, suggesting that crossing-over is a common outcome irrespective of the pathway utilised. This poses a problem. How can cells resolve joint molecules, such as Holliday junctions, to generate crossover products without cleavage-ligation? We suggest that the mechanism of bacterial DNA replication provides an answer to this question and that RecG can facilitate replication through Holliday junctions.