Lambda red recombination without electroporation

Lambda red recombination without electroporation

We are searching data for your request:

Forums and discussions:
Manuals and reference books:
Data from registers:
Wait the end of the search in all databases.
Upon completion, a link will appear to access the found materials.

All protocols that I found about Lambda red recombination are using Electroporation as a method to introduce (inject) the homologous DNA (usually a PCR products or a linear dsDNA) to the E.coli cell. Can I use chemical method (CalCl2 + PEG) and heat shock method instead ?

Since Lambda red recombination is a rare event you want to have to have a good transformation efficiency, thus electroporation would be your best choice. Also since the expression of the recombination system usually needs growing bacteria at 42º you will probably inhibit the expression of the recombination machinery with the chemical method.

Combining Methods of Gene Editing for Applications in Biotechnology

In 2012, I came across the concept of bacterial expressions systems within a practical setting, when Escherichia coli (E.coli) organisms were used to produce recombinant human tropoelastin, at the Department of Biochemistry, University of Sydney 1 . The synthetic protein tropoelastin is a precursor to the protein elastin, naturally present in the extracellular matrix (ECM) 2 subsequently used for bioengineering applications with biomaterials 3,4 . Recombineering is a genome engineering method primarily used within living E.coli bacteria, allowing easy manipulation of bacterial chromosome for metabolic engineering and production of a wide range of biochemicals 5 . Integration of systems biology, synthetic biology and evolutionary engineering has broadly enabled an assortment of genetic tools and protocols for efficient, cost-effective manipulation of E.coli in research and industrial applications 6 . To execute recombineering protocols, a basic knowledge of molecular and microbial techniques is required. This is a summary on current advancements of gene editing and its applications in synthetic biology.

New era of synthetic biology

At present, rapidly evolving gene technologies are anticipated to play a significant role towards addressing growing global challenges. Genetic engineering has revolutionized studies in molecular biology for the last 30 years since the discovery of restriction enzymes 7 . Restriction enzymes are an important tool in genomic research, akin to molecular scissors that cut DNA at a specific site and create a space to insert foreign DNA for editing purposes. Recombinant DNA technology, introduced in the 1970s created a pathway to insert foreign genes into E.coli host cells via shuttle vectors (a molecular vehicle) called plasmids, for gene expression and recombinant protein production (Figure 1). Owing to intrinsic limitations of plasmid-based applications including low yield of proteins, deleterious dosing of genes and significant metabolic burden to host cells, foreign genes were later integrated directly into the E.coli genome instead. Thus over the past decade, many advancements were made in gene engineering starting with the replacement of in vitro (in-lab) recombinant genetic engineering technology with in vivo (in organism) recombineering technology, for more precise, rapid and practically efficient results 8 . Importantly, the in vivo recombineering technology does not require restriction enzymes or DNA ligases, which were an essential component of classical genetic engineering 9 .

Figure 1: .Using restriction enzymes to cut a DNA fragment of interest and paste it with ligases into the plasmid "vectors". The pioneering work introduced in the early 1970s is still a widely used approach (Figure from AbFrontier).


Since its inception in 1998, recombineering or recombinogenic engineering has allowed direct engineering of bacterial chromosome within the organism (in vivo) by homologous recombination 9,10 . Homologous recombination is the process by which segments of DNA are exchanged between two DNA molecules through regions of identical sequence, to result in the new combination of genetic material 7 . Bacteria such as E.coli, encode their own homologous recombination systems, while viruses (or phages) that inhabit bacteria carry their own recombination functions and these two systems can work with or independently of each other 11 . Biotechnology has taken advantage of such phage proteins within bacteria to make use of homologous recombination for DNA engineering in E.coli, thus enabling a chosen DNA region to be cloned from a complex mixture, without restriction enzymes or prior isolation 11,12 . The most popular phage systems (virus systems inhabiting bacteria) are known as RecET based on genes encoded by RAC prophage and lambda Red, based on the homologous recombination system of bacteriophage λ (lambda), present in some E.coli strains 13 .

The segments of donor DNA used for recombineering can either be linear double stranded DNA (dsDNA) or synthetic single stranded oligonucleotides (ssDNA oligos) 13 , efficiently used in a variety of approaches (to create single-base changes, small deletions, large deletions and small insertions) and engineer DNA in E.coli 9 . The linear donor DNA substrate with desired change can be introduced by electroporation into bacterial strains to express the recombination functions, allowed to recover for a period of time and screened to identify recombinant colonies 5,8 . The phage-based recombination system can catalyze homologous recombination between linear DNA and a replicon such as phage λ, bacterial chromosomes, plasmids and bacterial artificial chromosomes (BACS) 9. Researchers strategically plan the recombineering pathway most appropriate to produce desired construct based on several different, extensive protocols in molecular biology 8 . Compared to classical genetic engineering, the editing process between a linear donor DNA substrate and replicon (BAC) is comparatively less labour-intensive and more efficient via recombineering (Figure 2).

Figure 2: Modifying the BAC clones in E.coli. a) In traditional genetic engineering, DNA fragments are cut and rejoined using restriction enzymes and DNA ligase, for classical homologous recombination, b) problems associated with the traditional approach have been overcome by the use of phage-mediated recombination in recombineering technology (Figure from reference 12 ).

Lambda Red Recombineering in Escherichia coli

Linear DNA introduced into bacterial systems are normally prone to degradation by bacterial nuclease enzymes 7 . Over the past decade, lambda Red recombineering has been used as a powerful technique for precisely defined chromosomal alterations in E.coli 10 . Three phage-derived lambda Red proteins: Gam, Exo and Beta are necessary to complete dsDNA recombination 7,10 Gam prevents degradation of foreign linear double stranded DNA by the E.coli nucleases, Exo degrades dsDNA to form a single stranded DNA (ssDNA) and Beta binding facilitates recombination (Figure 3). The exact mechanism on how a desired construct recombines with the chromosome in the presence of the three lambda Red proteins has been highly debated 14 . Furthermore, λ Red DNA-based engineering is the only method that allows multiplexing (sequential editing) in one step by direct electroporation, thereby used extensively as an editing tool for Multiplex Automated Genome Engineering (MAGE) 13 .

Figure 3: Overview of bacteriophage λ red recombination system used for recombineering. 1) Exo protein has exonuclease activity (5' to 3') generating 3' overhangs on linear DNA. 2) Beta binds the single-stranded DNA (3' overhangs), promotes ss-annealing and generates recombinant DNA. 3) An additional protein Gam (not shown here) prevents the E.coli enzymes from degrading double-stranded linear DNA fragments required for dsDNA recombineering (Figure from reference 5 ).

Coupling CRISPR/Cas9 system with Lambda Red Recombineering in E.coli

Despite novel techniques incorporated, the gene and protein yield with such bacterial expression systems remain small/insignificant leading to a bottleneck in metabolic engineering and systems biology. For instance, efficiency of short ssDNA/dsDNA oligonucleotide-mediated recombineering is highest for short genome modifications, while larger modifications occur with significantly lower frequency (<1%) 15 , necessitating screening via polymerase chain reaction (PCR an oligonucleotide amplification technique in molecular biology) to identify for the desired mutations that do not result in a clear phenotypic change 15 . Furthermore, to select for recombinant clones, recombineering methods rely on insertion of antibiotic markers to the gene, which must be removed prior to further modifications, thereby causing scars in the genome. Scars promote genome rearrangement after several rounds of modifications making it difficult to further engineer the strain. These limitations can be overcome by coupling lambda red recombineering with the currently powerful tool in molecular biology clustered regularly interspaced short palindromic repeat (CRISPR)/Cas9 system, which has proven ability to edit genes of a variety of organisms including Eukaryotes, Prokaryotes, higher plants and even human cell lines 13,15 . By function, Cas 9 is a programmable nuclease that can mediate a blunt double-stranded break (DSB) at almost any target DNA loci missing the desired mutation and since DSB is lethal, this will result in overall cell death thereby acting as a better selection/screening procedure for cells possessing intended modification generated via homologous recombination vs the wild-type (non-mutant cells) 13 . Since CRISPR/Cas9-coupled recombineering doesn’t therefore rely on antibiotic marker for selection, these gene replacement mutants are ideally “scar”-less 15 . The process is also performed in a shorter period of time than previous protocols.

In 2016, the technology was advanced one step further, to integrate automation (MAGE) for a combined CRISPR/Cas9 and λ Red recombineering-based MAGE technology, known as CRMAGE 15 . This system can be multiplexed to enable introduction of at least two mutations in a single round of recombineering with similar efficiencies, for multiple engineering rounds per day 15 . In a single round of CRMAGE recombineering, scientists were able to achieve close to 98% efficiency for a single point mutation which is unprecedented compared to previous efficiency of 5% using traditional MAGE. The study further demonstrated that it’s possible to modulate protein translation with relatively high frequency (from 6% to 70%), within the cell population as well 15 . Any regulatory element can thereby be inserted to modulate either gene expression or protein translation. Technically, it is even possible to obtain a clean, final recombinant strain by removing the entire CRMAGE editing system from the bacterial population after recombineering 15 . To speed up the design process for editing, the research team have facilitated a web-based tool that can predict the oligonucleotides for use in the editing process. The CRMAGE technique will enable generation of multiple mutations in a single cycle and multiple cycles within a single working day, to significantly increase the daily capacity of engineering the genome of E.coli and other bacterial strains, for potential high-throughput, research and industrial applications in biotechnology.

Poster Image: Screenshot from - 'What is genetic modification?' animated video of The Royal Society, available via YouTube.

Mini-lambda: a tractable system for chromosome and BAC engineering

The bacteriophage lambda (lambda) recombination system Red has been used for engineering large DNA fragments cloned into P1 and bacterial artificial chromosomes (BAC or PAC) vectors. So far, this recombination system has been utilized by transferring the BAC or PAC clones into bacterial cells that harbor a defective lambda prophage. Here we describe the generation of a mini-lambda DNA that can provide the Red recombination functions and can be easily introduced by electroporation into any E. coli strain, including the DH10B-carrying BACs or PACs. The mini-lambda DNA integrates into the bacterial chromosome as a defective prophage. In addition, since it retains attachment sites, it can be excised out to cure the cells of the phage DNA. We describe here the use of the mini-lambda recombination system for BAC modification by introducing a selectable marker into the vector sequence of a BAC clone. In addition, using the mini-lambda, we create a single missense mutation in the human BRCA2 gene cloned in a BAC without the use of any selectable marker. The ability to generate recombinants very efficiently demonstrates the usefulness of the mini-lambda as a very simple mobile system for in vivo genome engineering by homologous recombination, a process named recombineering.

Results and Discussion

Modification of the Red recombinase methods

Our first attempts at using the protocol described by Datsenko and Wanner [20] to construct recombinant prophages were unsuccessful. After several trials, we decided to modify the protocol. The modifications allowed the use of this methodology with the special characteristics of our strains, which carry inducible Shiga toxin-converting prophages.

The most obvious drawback was the spontaneous activation of the phage's lytic cycle during the process. This affirmation is supported by the isolation of phage DNA from the supernatants of the bacterial cultures without previous induction (data not shown). This increased prophage excision and/or phage release, which significantly reduced the efficacy of recombinant clone formation. Two possible causes of this failure were considered. Possibly, excision of the prophage DNA from bacterial DNA occurred without phage particle formation. In this case, the target gene (stx) where the amplimer must recombine would be lost, thus hindering correct recombination. Alternatively, after excision, phage particles were formed and released from the cells by lysis. In this case, a significant proportion of bacterial cells would be lysed, reducing the number of cells susceptible to transformation. Both of these scenarios would lead to a reduction in the efficacy of obtaining recombinant clones. The first cause could not be checked, but the second one was experimentally confirmed by the moderate reduction of the OD600 observed in the cultures of the lysogens during the process, compared with the OD600 of E. coli C600 or DH5α cultures (non-lysogens) used as a control.

The spontaneous induction of the lytic cycle could be due to several causes that would activate the SOS response. For example, the double application of the protocol to prepare electrocompetent cells (firstly to transform plasmid pKD46 and secondly to transform the PCR amplimer) which may activate the stress response in the bacterial cell. In any case, we assayed several modifications of the original protocol to optimize our application. These modifications of each step and the results obtained are described in this section.

Confirmation of the presence of the stx genes after pKD46 vector transformation

We examined the efficacy of the transformation of the helper plasmid pKD46 in the first step of the method, to evaluate the efficacy of our transformation protocol. Vector pBC-SK was used as a control. Vector pKD46 presented a lower transformation efficacy than pBC-SK. The number of transformed colonies was on average 6–10 fold higher with pBC-SK. These results could be due to the fact that pKD46 carries a temperature-sensitive origin of replication and only replicates at 30°C. Therefore the growth rate of cells transformed with pKD46 is lower than that of cells transformed with pBC-SK and grown at 37°C.

In addition to these expected results, the introduction of the pKD46 vector produced the loss of the stx 2gene in some of the lysogens. This was observed in a high proportion of the analyzed colonies (Table 1). The percentage of gene loss in such lysogens was significantly higher (t-Student, p < 0.05) than the percentage of stx 2gene loss in transformed cells after transformation with vector pBC-SK, used as control (Table 1). Only lysogens C600(933W) and C600(A9) did not show any gene loss which could suggest a higher stability of lysogens obtained with strain C600.

Since the stx 2gene was required to continue with the protocol, the presence of the vector and the stx 2gene in the selected clones was confirmed by a double hybridization with the respective stx and Red recombinase probes. Only clones in which the pKD46 and the stx 2gene were observed could be used in the next steps to transform the amplimer containing the antibiotic gene.

Other phage sequences, such as the rho independent terminator, the cI gene or the Q gene, which are present in the studied phages, were also absent in colonies lacking the stx 2gene (data not shown). This suggests that excision of the whole prophage DNA from the host cell occurred. However, since this was not the aim of this work, further investigations were not undertaken.

There is no explanation for the loss of the stx gene in a high percentage of the cells transformed with pKD46. Conditions were the same for all lysogens and some of them did not lose the stx. A possible hypothesis is that the presence of the Red recombinase system could increase the excision of certain prophages from the bacterial genome of some lysogens, without virus formation and without subsequent cell lysis.

Construction of the amplimer

One of the most important modifications to the protocol involved the construction of the PCR amplimer. In the first set of experiments, we used two primer sets to amplify the resistance marker containing each 36 bp and 40 bp extension sequences (short homologous arms) with homology to the stx genes (Table 2). These shared sequences were expected to be effective with the Red recombinase system. The primers were designed to obtain a Tc amplimer and a Cm amplimer that could be inserted into the recombinant phages. However, no antibiotic-resistant transformants of our lysogens were obtained using these primers. Therefore they were excluded from the experimental procedure.

The presence of Tc-resistant transformants was only observed on a few occasions when using the above described primers. However, the Tc cassette was not present inside the stx 2gene but somewhere else in the bacterial chromosome, despite the fact that all the plasmids used as templates of the antibiotic cassettes had conditional replicons to avoid erroneous integrations as suggested in the original protocol. Digestion of the amplimers with DpnI was to reduce this problem did not improve the results. The use of short homology sequences (36–40 bp) for the recombination could possibly be the cause of the wrong recombination events or the lack of recombinations. For this reason, the length of the homology regions where the recombinase can act, was increased. For this purpose a new strategy, based on the 3S-PCR protocol [29] was used to increase the length of the homology region shared by the amplimer and the stx 2gene.

The unusual recombination events are difficult to explain. Conceivably, the presence of the stx lambdoid prophages, some of which could include different recombinase genes, might somehow enhance non-homologous recombination. This would lead to the insertion of the antibiotic resistance cassette in unexpected sites outside of the target gene. In fact, the sequencing of several E. coli genomes has revealed the widespread occurrence of multiple integrases of phage origin [30] with similar sequence homology. These can help to explain our observations, which have also been described by other authors [31].

To increase the length of the stx 2homologous region upstream and downstream of the antibiotic resistance cassette, new amplimers were created by using overlapping regions between three different fragments: the 3' fragment, containing the stx 2homology the antibiotic resistance cassette and the 5' cassette, which also contains the stx 2homology (Fig 1). Some amplimers, such as the one containing tet, were directly obtained using the three fragments simultaneously as templates in the same PCR reaction. For cat, the 3'-fragment joined to the antibiotic cassette was amplified in a first PCR reaction and the antibiotic cassette joined to the 5'-fragment in a second one. Then both fragments were used together to obtain the complete amplimer in a third PCR reaction.

Scheme of the protocol used to construct the amplimers containing the stx gene substituted by the antibiotic resistance gene. The figure shows the protocol used for the cat gene as an example.

The use of longer homologous regions (280 bp at each side) in the amplimer carrying the antibiotic cassettes definitely solved the problem of false recombinants (Table 2). In the above mentioned application (29), the rational for the use of longer regions of homology was to increase the probability of the double event of recombination that lead to the desired allelic exchange in non-E. coli K-12 bacteria. This was explained by the authors since the lambda-red functions became less efficient in blocking DNA degradation in bacteria that are distantly related to E. coli. Therefore, longer regions of homology increased the efficiency. Although our application was performed using an E. coli K-12 derivative, strain DH5α, the same strategy of long arms seems to be suitable for our purposes.

The amount of the amplimer to be transformed was also evaluated and finally established at 0.5 μg of amplified DNA (Table 2). From 10–100 ng of amplified DNA were established in the original protocol [20]. However, in our hands and using an amplimer longer than the one described in the original protocol, higher concentrations were necessary.

Transformation of the amplimer

Initially, we used 5 ml SOB cultures with ampicillin and 10 mM of L-arabinose as described by Datsenko and Wanner [20]. However, no recombinant clones were obtained. In successive attempts, lysogenic cells (pKD46 + , stx 2 + ) in culture volumes of 10, 25 and finally 50 ml were used to prepare electrocompetent cells (Table 2). Electrocompetent cells prepared from 50 ml of culture were finally used to obtain recombinant clones. This problem did not apply in the approach described by Datsenko and Wanner [20], but in our approach it was necessary to obtain few recombinant colonies. This confirmed our hypothesis that in our cultures the initial number of cells was probably reduced by the activation of phage lysis during the protocol for preparation of electrocompetent cells or by the presence of antibiotic in the culture media. Therefore the minimal numbers of cells in the initial culture necessary to obtain a single recombinant colony was of 2.5 × 10 10 CFU/ml.

Different concentrations of L-arabinose were also tested. In optimal conditions, no significant differences were observed in our experiments when using different amounts of arabinose. We assayed 1 mM (indicated in the original protocol), 10 mM and 0.1 M of L-arabinose. Recombinant clones were obtained in all cases. However, the highest number of clones was obtained with 0.1 M of arabinose (Table 2). These results indicate that all of the concentrations tested were inside the range needed to generate the expression of the γβexo gene system in a proportion of the transformed cells. This is required in a system that is dependent on arabinose concentrations and that has been described as "all-or-none induction of PBAD" [32].

After transformation of the amplimer, the cells were recovered in 1 ml of SOC medium and incubated for 1 to 3 hours at 37°C (as suggested in the original method) or 30°C before plating. Three hours of incubation were required to obtain recombinants. Although pKD46 is supposed to be a temperature sensitive vector and cannot replicate above 30°C, the proteins encoded by the vector still can be active inside the competent cells enhancing recombination. There were no great differences in the number of clones obtained from incubation at different temperatures (Table 2) although more colonies were obtained at 37°C. In fact, incubation at 37°C increased the growth rate, producing more cells.

Plating media

LB agar plates containing Tc and Cm were used as selective media. Datsenko and Wanner [20] proposed the use of 25 μg/ml for Cm and Km. Initially, we tested concentrations of 20 μg/ml Tc or 20 μg/ml Cm for the recovery of recombinants. However, no recombinants were observed on Cm or Tc plates (Table 2). Positive results were obtained when using lower antibiotic concentrations in the selective media. Therefore, LB plates containing 5 μg/ml Tc or 5 μg/ml Cm were finally used for the recovery of recombinants. Once plated, 24 hours of incubation were enough to visualize colonies. The colonies were then transferred to new LB agar plates with higher antibiotic concentrations (Tc: 20 μg/ml and Cm: 20 μg/ml respectively). Thus lysogens carrying recombinant phages did not grow immediately in the expected antibiotic concentrations, possibly be because bacterial cells need to recover after the process, and were damaged when directly submitted to high antibiotic concentrations.

Loss of the vector was achieved after successive subcultures and incubation at 43°C without ampicillin selection, as previously described [20]. The loss of the plasmid was confirmed by observing the absence of the PCR amplimer, using the primers described in Table 1 for vector pKD46.

Table 2 summarizes all the variations assayed in our protocol the number of recombinant clones obtained with two of the bacteriophages using the different modifications and the final conditions that were established.

Recombinant phages

The tet and cat genes were introduced in the stx 2gene of prophages: ØA9, ØA312, ØA534, ØA549, ØA557, ØVTB55 and 933W. The genes were placed at position 251 bp of the stx 2-A subunit and 264 bp upstream of the end of the stx 2-B subunit. The identification and characterisation of the respective gene in each recombinant prophage was achieved by PCR and sequencing.

The induction of stx 2-phages from the six recombinant lysogens had slightly different kinetics than the original lysogens. Nevertheless, all recombinant lysogens conserved their lytic capacity after incorporating the antibiotic resistance gene. As previously described [33, 34], plaques produced by stx-phage are usually poorly visible or turbid on the top agar layer. Therefore, confirmation of the presence of infective recombinant phages was achieved by plaque blot hybridization with specific cat or tet probes (data not shown).

The antibiotic resistance cassette appears to remain stable within the phage genome, as observed after four steps of subculture without antibiotic selection. However the long-term stability of the marker gene in the phage genome without antibiotic selection remains to be elucidated.

Recombinant transductants

To evaluate the capacity of the recombinant phages to infect and convert new E. coli strains, suspensions of the recombinant phages were prepared. Transduction of E. coli DH5α and E. coli C600 was performed. All phages produced new transductants, which conferred resistance to the appropriate antibiotic on DH5α or C600. Presence of the recombinant prophages in the host strain was confirmed by PCR and plaque hybridization analysis (Figure 2).

A) Number of transductants (CFU/ml) obtained in E. coli DH5α and E. coli C600 with each recombinant phage carrying ca t or tet antibiotic resistance genes. B) PCR products of each recombinant colony carrying a stx-prophage with the respective antibiotic resistance cassette. 1: stx gene control, 2: cat control, 3: tet control, 4–5: E. coli DH5α and E. coli C600 lysogens with stx-phages::cat. 6–7: E. coli DH5α and E. coli C600 lysogens with stx-phages::tet-. 8 negative PCR control C) As an example, colony blot of DH5α (Ø557::tet) hybridized with the specific probe for the tet gene.

Some other authors have produced recombinant stx-phages [9, 15, 16] for different purposes. Schmidt et al. [9] used phage φ3538 to infect and lysogenize enteric Escherichia coli strains and to develop infectious progeny from such lysogenized strains. Allison et al., [15] used recombinant phages to show the first reported observation of the simultaneous infection of a single host with two genetically identical Stx phages. Acheson et al., [16] generated a recombinant Shiga toxin 1-converting phage H-19B to facilitate the study of intestinal transmission of stx1-phages.

In the present work we have generated different recombinant phages with two antibiotic resistance genes to use them for different purposes. The use of these phages will allow analyse transduction in different matrices, as food or water samples. It would also be interesting to evaluate the phage induction and the transduction after different processes applied for food or water treatments, such as high temperature or high hydrostatic pressure (HHP), which has been reported as a method that can generate an increase in the induction of the lytic cycle of certain stx-phages [35].

The recombinant phages would be also useful tools to evaluate the ability of stx-phages to generate double lysogens and to evaluate whether the double lysogeny is really favoured in STEC, as some observations done in water environments or in strains isolated from humans and animals would suggest [15, 33, 36, 37].

High-efficiency scarless genetic modification in Escherichia coli by using lambda red recombination and I-SceI cleavage.

Genetic modifications of bacterial chromosomes are important for both fundamental and applied research. In this study, we developed an efficient, easy-to-use system for genetic modification of the Escherichia coli chromosome, a two-plasmid method involving lambda Red (λ-Red) recombination and I-SceI cleavage. An intermediate strain is generated by integration of a resistance marker gene(s) and I-SceI recognition sites in or near the target gene locus, using λ-Red PCR targeting. The intermediate strain is transformed with a donor plasmid carrying the target gene fragment with the desired modification flanked by I-SceI recognition sites, together with a bifunctional helper plasmid for λ-Red recombination and I-SceI endonuclease. I-SceI cleavage of the chromosome and the donor plasmid allows λ-Red recombination between chromosomal breaks and linear double-stranded DNA from the donor plasmid. Genetic modifications are introduced into the chromosome, and the placement of the I-SceI sites determines the nature of the recombination and the modification. This method was successfully used for cadA knockout, gdhA knock-in, seamless deletion of pepD, site-directed mutagenesis of the essential metK gene, and replacement of metK with the Rickettsia S-adenosylmethionine transporter gene. This effective method can be used with both essential and nonessential gene modifications and will benefit basic and applied genetic research.


MAGE is a powerful technique that can be used to generate combinatorial sets of designed mutations in a population (4) and/or modify hundreds of alleles in a single strain (5). We have engineered optimized strains for multiplex genome engineering in an effort to streamline extensive genome editing. Previously, we showed that converting a selectable allele in the vicinity of multiple non-selectable alleles enriches the candidate pool for highly modified clones (9). Additionally, we demonstrated that exonucleases are capable of degrading single-stranded MAGE oligos even when these oligos are protected using phosphorothioate bonds (Mosberg, J.A., Gregg, C.J., et al., in review). Inactivating ExoI, ExoVII, ExoX, RecJ and 㮾xo significantly enhanced multiplex AR frequencies (Mosberg, J.A., Gregg, C.J., et al., in review). This showed that intracellular MAGE oligos are a limiting factor in Redβ-mediated recombination. In the current work, we demonstrate that available ssDNA on the lagging strand of the replication fork is another limiting factor that can be increased by disrupting the interaction between DnaG primase and DnaB helicase on the replisome.

In order to increase ssDNA on the lagging strand of the replication fork, we introduced two known mutations in primase (DnaG)—K580A and Q576A. These mutations have been shown in vitro to increase OF size by interrupting the primase–helicase interaction on the replisome (13). Based on the measurements of Tougu et al. (13), we estimate that the K580A mutation increases OF length by 𢏁.5-fold and the Q576A mutation increases OF length by 𢏈-fold (Supplementary Table S2). EcNR2.dnaG.K580A and EcNR2.dnaG.Q576A exhibited significant increases in the mean number of alleles converted and decreases in the proportion of clones with zero non-selectable alleles converted. Furthermore, the strongest enhancement was observed in EcNR2. dnaG.Q576A (the variant with the longest OFs of the strains reported herein), with an intermediate enhancement observed in EcNR2.dnaG.K580A (the variant with intermediate-sized OFs). This relationship between recombination frequency and OF length further supports the model in which Redβ mediates annealing at the lagging strand of the replication fork (3,15,16,19), and our hypothesis that ssDNA on the lagging strand of the replication fork is a limiting factor during this process. With this in mind, we unsuccessfully attempted to generate a DnaG Q576A/K580A double mutant, suggesting that such an extensive manipulation of the DnaG C-terminal helicase interaction domain (24) was lethal.

Our results indicate that intracellular concentrations of MAGE oligos and the accessibility of their genomic targets are both limiting. To further increase the number of simultaneous mutations that can be generated by CoS-MAGE, it is helpful to understand whether the AR frequency is limited predominantly by the number of oligos that enter the cytoplasm, or whether kinetics are also relevant. Since a maximum of 9 ARs was observed for the 10-oligo sets compared to a maximum of just 12 ARs for the 20-oligo set, oligo uptake may be limiting. However, the fact that primase modulation—in addition to nuclease inactivation𠅎nhances AR frequency underscores the kinetic constraints regarding Redβ-mediated annealing. Each missed opportunity to anneal (i) increases the number of wt alleles in the population due to replication and (ii) decreases the number of MAGE oligos available, via dilution (cell division) and degradation (nucleases). Increasing the concentration of each reactant (i.e. intracellular oligos and accessible genomic targets) would increase the kinetics of annealing. Therefore, the number of intracellular oligos may limit the maximum number of possible mutations, but kinetics appear to be a significant force limiting the population-wide AR frequency average.

Interestingly, the nuclease-deficient Nuc5- strain (Mosberg, J.A., Gregg, C.J., et al., in review) performed statistically similarly to the EcNR2.dnaG.Q576A strain for Sets 1 and 2, whereas EcNR2.dnaG.Q576A strongly outperformed the nuclease-deficient strain for Set 3 ( Tables 1 and ​ and2 2 see also Mosberg, J.A., Gregg, C.J., et al., in review). While oligo design parameters such as type of designed mutation (4), oligo length (4), oligo secondary structure (4) and off-target genomic homology (5) are major determinants of AR frequency, our results highlight the relevance of genomic context. This has previously been difficult to demonstrate, but is apparent from the discrepancy in performance of the same oligo sets tested in our Nuc5- (Mosberg, J.A., Gregg, C.J., et al., in review), EcNR2.dnaG.Q576A, and Nuc5-.dnaG.Q576A strains. For example, different regions may have different replication fork speed or priming efficiency. These factors could locally modulate OF length, thus affecting Redβ-mediated AR frequency (although replication fork speed did not appear to be a major factor in vitro (13)). Therefore, increasing the region that must be replicated by a single OF may profoundly increase AR frequency for oligos targeting such regions. Alternatively, certain oligos may be more susceptible to nuclease degradation, so removing the responsible nucleases would disproportionately improve AR frequency for such oligos. With this in mind, we tested whether combining primase modification and nuclease removal would enhance MAGE performance more than either strategy used individually. Indeed, Nuc5-.dnaG.Q576A consistently performed the best ( Figures 3 and ​ and5) 5 ) of all tested strains. Therefore, the two disparate strategies can be combined for a larger and more robust MAGE enhancement.

To explore the extent to which OF localization impacts CoS-MAGE performance, we tested whether placing 10 oligos within a single putative OF would yield subpopulations of unmodified (few alleles converted) and ‘jackpot’ (most alleles converted) recombinants. However, CoS-MAGE using the densely clustered lacZ oligos ( Figure 4 ) produced a similar AR distribution to the ones observed for Sets 1𠄳 ( Figure 3 ), which target regions of the genome spanning several putative OFs. Since mutations within a single putative OF behaved similarly to mutations spread across many OFs, nascent OF placement does not appear to be a critical determinant of multiplex AR frequency. A number of hypotheses could explain why the expected ‘jackpots’ are not observed. Most likely, MAGE oligos are limiting due to degradation and/or lack of uptake. Thus, it is possible that most cells lack some of the oligos necessary for generating a majority of the desired mutations. Additionally, OF extension may occur too fast for all of the MAGE oligos to anneal before the OF occludes their targets. Still another explanation could be that ssDNA binding proteins occlude ssDNA on portions of the lagging strand, rendering these regions non-accessible for Redβ-mediated annealing. Finally, it is also possible that several MAGE oligos annealed within a single OF could destabilize lagging strand synthesis, leading to selection against highly modified ‘jackpot’ clones. Indeed, Corn and Berger (25) hypothesize that DnaG primase has evolved to only initiate synthesis when multiple DnaG units are bound to DnaB Helicase, as OF synthesis away from the replisome could be detrimental. Since polIIIlag dissociates from the replisome after completing an OF (26), the rapid and repeated dissociation of polIIIlag caused by multiple nearby MAGE oligos could inhibit lagging strand synthesis as the replisome proceeds beyond the target region. In the absence of the rest of the replisome, a cytosolic PolIII holoenzyme alone can synthesize 1.4 kb on a ssDNA template primed by 30 nt DNA oligos (27), but this activity is considerably diminished compared to that of an intact replisome. Therefore, if OFs are not completed when the replisome is in close proximity, this could result in persisting ssDNA that could destabilize the chromosome and/or cause lesions when the next replication fork passes through.

We also investigated whether targeting a greater number of alleles would increase the resulting number of conversions in our enhanced strains ( Figure 5 ). Although the mean number of alleles converted (mean ± std. error of the mean) increased from 2.59 ± 0.19 with 10-oligo Set 3 to 4.50 ± 0.30 (1.74-fold) with 20-oligo Sets 3 + 3X for Nuc5-.dnaG.Q576A, the mean number of alleles converted for EcNR2.dnaG.Q576A only increased from 2.54 to 2.96 (1.17-fold). The superior enhancement for the nuclease-depleted Nuc5-.dnaG.Q576A strain suggests that the intracellular oligo concentration is a limiting factor for highly multiplexed MAGE (㸐 alleles targeted). Therefore, enhancing DNA uptake and/or preservation may be a fruitful means of further improving MAGE. However, the greater multiplexibility of Nuc5-.dnaG.Q576A could also be due to the 10 new Set 3X oligos being more responsive to decreased exonuclease degradation than to increased lagging strand ssDNA availability. Additionally, there may be other limiting factors such as insufficient Redβ or unidentified host proteins. Although there is no known precedent for limiting amounts of λ Red proteins during recombination (28), our novel ability to attain 12 simultaneous non-selectable ARs ( Figure 5 A) shows that our improved strains are in uncharted territory for probing the limits of λ Red recombination.

Given that DnaG primase acts solely on the lagging strand of the replication fork, we expected that the primase modifications would only enhance lagging strand recombination. Therefore, the performance of leading-targeting CoS-MAGE in our strains was surprising, as EcNR2.dnaG.Q576A significantly outperformed EcNR2 (*p = 0.018). Furthermore, while the total number of tolC+ recombinants was far smaller (� 2 -fold) for leading-targeting CoS-MAGE, the AR frequency of non-selectable alleles in these recombinants was still quite impressive, especially in extremely close proximity to the selectable allele. This suggests that one leading strand recombination event strongly correlates with multiple additional recombinations. Two possible explanations for the superior performance of EcNR2.dnaG.Q576A in leading-targeting CoS-MAGE are that (i) an impaired primase–helicase interaction increases accessible leading strand ssDNA or (ii) infrequent Redβ-mediated strand invasion initiates a new replication fork that travels in the opposite direction and swaps which strand is the lagging strand.

There is strong support for primase function affecting the dynamics of replication on both the lagging and leading strands (26,27,29). Lia et al. (26) observed phases in which OF synthesis is faster than helicase progression at the replication fork, alternating with phases in which helicase progression outstrips the rate of OF synthesis by PolIIIlag. These results demonstrate that DnaB-PolIIIlead does not progress at the same instantaneous speed as PolIIIlag (26). Furthermore, Yao et al. (29) showed that the velocity of leading-strand synthesis decreases during lagging strand synthesis, while its processivity increases. Perhaps less frequent primase–helicase binding leads to transient asynchrony of the helicase and PolIIIlead. Given that PolIII tends to release from the replication fork more readily than does DnaB helicase (29), a transiently increased fork rate and decreased PolIIIlead processivity could exacerbate such an asynchrony, creating a leading strand trombone loop similar to those observed during lagging strand synthesis. However, the effects of lagging strand synthesis on leading strand replication have been historically difficult to demonstrate in experiments beyond single-molecule studies (29). Given that instantaneous changes in replication dynamics appear to occur on timescales relevant to Redβ-bound oligo recombination, it is conceivable that snapshots of exposed ssDNA on the leading strand template could be recorded by measuring rates of leading-targeting AR. Single-molecule analysis of the Q576A variant could explore this hypothesis.

Alternatively, Redβ has been reported to facilitate strand invasion in vitro (30). If this also occurs in vivo, such strand invasion would produce a D-Loop that could act as a new origin of replication (31). Therefore, invasion of one leading–targeting MAGE oligo could initiate a replication fork traveling in the opposite direction. In the reverse orientation, the leading strand would become the lagging strand so that upstream oligos would become lagging targeting and much more likely to recombine. This could lead to the highly modified clones that we observed during leading-targeting CoS-MAGE (Supplementary Figure S1). If this is the case, the non-selectable alleles would be upstream of the tolC selectable marker. Since co-selection is most effective downstream of the selectable marker (9), this may explain why co-selection enhancements decay rapidly with distance on the leading strand.

In this manuscript, we have identified available ssDNA on the lagging strand of the replication fork as a limiting factor in multiplex genome engineering. Compared with a standard recombineering strain (EcNR2), EcNR2.dnaG.Q576A displays on average 62% more alleles converted per clone, 239% more clones with 5 or more allele conversions and 38% fewer clones with 0 allele conversions in a given round of CoS-MAGE with 10 synthetic oligos ( Table 2 ). We used this strategy to build on our recent advances (Mosberg, J.A., Gregg, C.J., et al. in review), generating the Nuc5-.dnaG.Q576A strain, which has extended OFs and also lacks five potent exonucleases. These modifications exploited two distinct mechanisms that together increased the robustness and potency of CoS-MAGE, enabling an average of 4.50 and a maximum of 12 ARs in single cells exposed to a pool of 20 different synthetic AR oligos ( Figure 5 ). Additionally, 48% of recombinants had 5 or more ARs and only 8% had 0 modified non-selectable alleles. Furthermore, in a given round of CoS-MAGE with 10 synthetic oligos, Nuc5-.dnaG.Q576A displays on average 111% more alleles converted per clone, 527% more clones with 5 or more allele conversions and 71% fewer clones with 0 allele conversions in comparison with EcNR2 ( Table 2 ). This improvement in MAGE performance will be highly valuable for increasing the diversity explored during the directed evolution of biosynthetic pathways (4) and for enabling the rapid generation of desired genotypes involving tens to hundreds of ARs (5).

High-Efficiency Scarless Genetic Modification in Escherichia coli by Using Lambda Red Recombination and I-SceI Cleavage

FIG 1 Diagram of the two-plasmid method. (A) Antibiotic cassette fragment with I-SceI recognition sites integrated at a target via λ-Red-mediated recombination, creating an intermediate strain. (B) The intermediate strain, with a mutation(s) and I-SceI recognition sites, was transformed with the donor plasmid. Expression of I-SceI was induced by l -rhamnose or IPTG. I-SceI recognition sites in the donor plasmid and the chromosome were cleaved. Integration of the donor fragment at the cleaved site of the chromosome was mediated by λ-Red recombination. FIG 2 Plasmids used for this method. (A) pREDTKI and pREDTAI (helper plasmids), with arabinose-inducible (araB promoter) λ-Red recombinase functions and IPTG-inducible (trc promoter) I-SceI expression. (B) pKSI-1 (modular plasmid for donors), based on the high-copy-number vector pBluescript II KS(−), with an MCS and two I-SceI recognition sites. (C) Part of plasmid pMDIAI (template plasmid), with the apramycin resistance gene flanked by FRT sites and I-SceI recognition sites. For pMDISI, the apramycin resistance gene was replaced with the spectinomycin resistance gene. Pink arrow, binding sites for primers MDF and MDR.
ModificationHelper plasmid/promoter for I-SceI expressionDonor plasmid backboneMarker eviction rate (no. of Apr- and/or Spc-sensitive colonies/no. of tested colonies)Donor plasmid curing rate (no. of Amp- or Kan-sensitive colonies/no. of tested colonies)Correct modification rate, confirmed by PCR (no. of colonies with expected PCR bands/no. of tested colonies)Correct modification rate, confirmed by PCR fragment sequencing (no. of colonies with expected sequence/no. of tested colonies)
cadA deletionpREDIA/rhaBpBackZero-T4/103/43/33/3
gntT integrationpREDTKI/trcpMD19-T8/107/87/74/4
pepD seamless deletionpREDKI/rhaBpKSI-13/82/32/22/2
metK mutationpREDKI/rhaBpKSI-19/305/99/94/4
metK replacementpREDTKI/trcpKSI-12/42/22/22/2

Markerless knockout and knock-in of selected genes. (i) Knockout of cadA.

(ii) Knock-in of gdhA.

(iii) Multiple modifications.

Seamless deletion of pepD.

FIG 3 Seamless pepD deletion and PCR analysis. (A) Construct with 40-bp short arms homologous to the pepD ORF and a 600-bp pepD segment replaced by a marker. (B) Construct with arms with upstream (U) and downstream (D) homology to pepD, with truncated 1.0-kb pepD segments and the apramycin (apr) resistance gene seamlessly deleted, from ATG to TAA. (C) PCR analysis of pepD expression. Band sizes: BL21(DE3) (wild-type pepD), 2.5 kb intermediate strain (pepD::apr), 3.2 kb final strain (ΔpepD), 0.9 kb.

Modifications in the essential metK gene.

FIG 4 Modifications to metK gene and PCR analysis. (A) Construct with IsceI-apr-IsceI cassette inserted into yqgC and IsceI-spc-IsceI cassette inserted into galP. yqgC and galP are nonessential genes next to metK. (B) Wild-type (WT) metK and resistance genes were replaced by mutated metK or the SAM transporter gene. The galP fragment and speA-yqgB-yqgC fragment were homologous arms. (C) PCR analysis of metK. Band sizes: MG1655 (wild-type metK), 2.8 kb intermediate strain A10 (yqgC::apr galP::spc), 5.9 kb final strain with mutated metK, 2.8 kb (the same length as wild-type metK) final strain with metK replaced by SAM transporter gene, 2.6 kb.

Recombineering with overlapping single-stranded DNA oligonucleotides: testing a recombination intermediate

A phage lambda-based recombination system, Red, can be used for high-efficiency mutagenesis, repair, and engineering of chromosomal or episomal DNA in vivo in Escherichia coli. When long linear double-stranded DNA with short flanking homologies to their targets are used for the recombination, the lambda Exo, Beta, and Gam proteins are required. The current model is: (i) Gam inhibits the host RecBCD activity, thereby protecting the DNA substrate for recombination (ii) Exo degrades from each DNA end in a 5' --> 3' direction, creating double-stranded DNA with 3' single-stranded DNA tails and (iii) Beta binds these 3' overhangs to protect and anneal them to complementary sequences. We have tested this model for Red recombination by using electroporation to introduce overlapping, complementary oligonucleotides that when annealed in vivo approximate the recombination intermediate that Exo should create. Using this technique we found Exo-independent recombination. Surprisingly, a similarly constructed substrate with 5' overhangs recombined more efficiently. This 5' overhang recombination required both Exo and Beta for high levels of recombination and the two oligonucleotides need to overlap by only 6 bp on their 3' ends. Results indicate that Exo may load Beta onto the 3' overhang it produces. In addition, multiple overlapping oligonucleotides were successfully used to generate recombinants in vivo, a technique that could prove useful for many genetic engineering procedures.


Model for Red-mediated recombination. λ…

Model for Red-mediated recombination. λ Exo enters a dsDNA end and degrades one…

Model to explain the high…

Model to explain the high recombination frequency of dsDNA with 5′ overhangs. The…


Essential bacterial genes are an especially interesting target for DMS because they play an important role in bacterial evolution (Long et al, 2015 Maddamsetti et al, 2017 ), the emergence of antibiotic resistance (Walsh, 2000 Allen et al, 2010 ), and strain engineering (Winkler et al, 2016 de Jong et al, 2017 ). It is important to modify the essential genes in their native genomic context. Expressing essential genes on plasmids alters the cellular fitness because of different expression levels due to copy number effects (Gibson et al, 2013 ) and the loss of epigenetic regulation. Current genome mutagenesis techniques suffer from low-editing efficiencies and mutational biasing, which greatly decrease the quality of the fitness data (i.e., due to the overabundance of wild type or a few over-represented members). As such, comprehensive DMS of essential genes using these approaches has remained elusive, especially in bacteria.

Bacterial genome-editing technologies have advanced greatly in the past decade. One technology is multiplexed automated genome engineering (MAGE) that is based on lambda Red-mediated recombination of single-stranded oligos for desired allelic exchange on the genome or “Recombineering” (Wang et al, 2009 ). The efficiency of introducing a single nucleotide change using recombineering is often very low and context-dependent (Sharan et al, 2009 ). Using MAGE, efficiency and throughput are improved by re-transforming the same large pool of oligos over repeated cycles (Wang et al, 2009 ). This technique has been successfully used for several outstanding synthetic biology and metabolic engineering applications (Wang et al, 2009 Sandoval et al, 2012 Lajoie et al, 2013b Raman et al, 2014 Amiram et al, 2015 ). However, it may be difficult to apply MAGE for DMS of essential genes. Due to the repeated transformation cycles during library construction, the mutations that are deleterious or even neutral to the host would be lost (Wang et al, 2009 ). Repeated heat shock and electroporation during the recombineering cycles in the MAGE protocol (Wang et al, 2009 ) would also add additional stress that would be detrimental to the diversity in essential genes. Additionally, in order to increase the efficiency of recombination in MAGE, significant modifications such as deletion of methyl-directed mismatch repair (MMR) and DNA primase (dnaG) are required, which change the native genetic context (Wang et al, 2009 Sawitzke et al, 2011 ). Deletion of mismatch repair systems increases the background mutation rate (Isaacs et al, 2011 ), which may also confound fitness estimates.

Genome-editing technologies developed using CRISPR/Cas9-mediated recombineering have helped address several challenges with MAGE. A chimeric guide RNA (gRNA) programs the Cas9 endonuclease to induce a DNA double-strand break (DSB) at any genomic target upstream of a 5′-NGG-3′ PAM sequence and complementary to the 20-bp spacer sequence in the gRNA (Jinek et al, 2012 ). In several bacteria, including Escherichia coli, the Cas9:gRNA-mediated DNA DSB induces cell death due to a lack of adequate DSB repair pathways (Jiang et al, 2013 ). Therefore, Cas9:gRNA-induced DSBs can select for PAM substitutions introduced by recombineering (Cong et al, 2013 Jiang et al, 2013 ). Synonymous PAM-inactivating mutations (SPMs) can be coupled to other mutations in the same recombination template for precise genome manipulation with high efficiency using a single transformation step (Pyne et al, 2015 Reisch & Prather, 2015 Bassalo et al, 2016 Chung et al, 2017 Wang et al, 2018 ).

High-throughput genome editing with Cas9-mediated recombineering was achieved recently using CRISPR-enabled trackable genome engineering (CREATE) (Garst et al, 2017 ). Using CREATE, the DNA encoding the gRNA expressed under a constitutive promoter was covalently linked to the DNA repair template on 250-bp editing cassettes (Garst et al, 2017 ). Over 100,000 editing cassettes can be synthesized on microarray chips and subsequently cloned in high-throughput into cells with active Cas9 and lambda Red recombination to generate genome-wide mutation libraries. The editing cassettes on the plasmid also serve as the barcode to track the mutations before and after selection to assign fitness scores to each mutation (Garst et al, 2017 ). The technology has been used for directed evolution of E. coli proteins, pathways, and strains (Shalem et al, 2014 Cobb et al, 2015 Cho et al, 2017 Liang et al, 2017 Liu et al, 2017 Lu et al, 2017 Wu et al, 2017a , b Zhu et al, 2017 Bassalo et al, 2018 ). However, applying CREATE for DMS proved to be challenging. Anywhere between 10 and 60% of randomly chosen gRNA targeting different genomic loci have been shown not to induce Cas9:gRNA-induced cell death (Cui & Bikard, 2016 Zerbini et al, 2017 ). Consequently, due to variable selection, editing efficiency can vary between 0 and 100% across gRNAs (Garst et al, 2017 Zerbini et al, 2017 ). Cells with gRNAs that fail to induce DSB-mediated cell death can grow significantly faster than cells with active gRNAs undergoing DSBs and editing (Jiang et al, 2015 Cui & Bikard, 2016 ). Consequently, in high-throughput non-DSB-inducing gRNAs, with low-editing efficiency, take over the population and reduce overall editing efficiency to only

1–4% (Bassalo et al, 2018 ). Several gRNAs also cause unintended mutations on the genome that are not encoded in the repair template (Cui & Bikard, 2016 Zerbini et al, 2017 ). Consequently, cells with no edits and unintended mutations can be falsely tracked as beneficial mutations. Finally, each gRNA is coupled to a different synonymous PAM mutation (SPM) and synonymous mutations can lead to significant fitness effects, especially in essential genes (Lind et al, 2010 Agashe et al, 2013 Lajoie et al, 2013a ). Because of these limitations, CREATE experiments have largely been limited to finding mutants with large fitness effects in the presence of strong selective pressures (Bassalo et al, 2018 Pines et al, 2018 ).

We posited that in order to target a single genomic locus, we could use a single pre-screened gRNA and synonymous PAM-inactivating mutations (SPM). In this study, we discuss the CRISPR/Cas9-mediated genomic error-prone editing (CREPE) technology. As opposed to other Cas9-mediated high-throughput technologies in E. coli, in the CREPE protocol we use a single gRNA to integrate an error-prone PCR library of the target with the SPM on the genome (Fig 1). Recently, a similar technology, CASPER, was reported in yeast (Jakočiūnas et al, 2018 ). However, yeast has a significantly higher recombination efficiency than bacteria such as E. coli. Recombination efficiency with linear dsDNA templates is very low in E. coli (Murphy et al, 2000 ), and recombineering using dsDNA template with limited single nucleotide changes is poorly understood. Therefore, we varied the homology arm length and the Cas9 recombineering system to improve recombination and our understanding of recombination using a repair template with single nucleotide changes. We successfully developed a platform that efficiently generates unbiased and diverse genomic mutant libraries with > 80% editing efficiency for non-essential genes and > 55% efficiency for essential genes. Additionally, while CASPER was used for directed evolution, we adapted CREPE for use as a DMS platform to study essential E. coli genes in their native genomic context. Using CREPE, we scored the fitness of naturally accessible mutations in the RNA polymerase beta subunit that confer resistance to rifampicin.


We demonstrated previously that Red-mediated recombination with synthetic single-strand oligos is very efficient and independent of RecA in E. coli. Only λ Beta appears to be required for this ssDNA recombination (Table 3) (9, 28, 36). Oligos that correspond to either of the two complementary DNA strands generate recombinants, but invariably one oligo recombines more efficiently than the other. By testing six markers in different regions of the chromosome, a pattern emerged (9). At each position, the most efficient of the two complementing oligos was the one corresponding to the lagging-strand DNA (i.e., the same sequence as the Okazaki fragments). We proposed that oligo-directed recombination occurred at the replication fork, and that the “lagging-strand oligo” is more easily annealed by Beta because of larger gaps present in the lagging strand (9, 29). Three points from the results presented here bear on our previous proposal that Beta-mediated recombination with ssDNA oligos occurs at the replication fork. First, we have demonstrated that four different oligos recombine more efficiently when each is targeted to the lagging strand than when targeted to the leading strand (Table 4). Second, we show that MMR can depress the appearance of recombinants by >100-fold (Table 2). The MMR functions MutS, MutL, MutH, UvrD helicase, and Dam methylase are required for mismatch repair at the replication fork (37). Third, when recombination occurs without bias due to MMR efficiencies, we see incredibly high recombination frequencies of 25%. It is difficult to come up with other mechanisms that generate a single-strand gap at a specific site in 25% of the cells during the time period of the experiment. We do not yet understand why 75% of the cells are resistant to recombination despite saturating levels of oligo. Among several possibilities, some cells may not be electrocompetent, may be in a state resistant to recombination, or may have the target sequestered from the oligo.

Although the lagging-strand oligos generate more recombinants than the leading-strand oligo, the leading-strand oligos are still very recombination proficient. Comparing four oligos (Table 4), lagging-strand recombinants are on average 30-fold more frequent than the leading strand. This may be explained by a proportionally different amount of ssDNA generated during replication on each side of the fork. By this logic, the gaps on the lagging-strand side would be 30 times greater than the gap on the leading-strand side. Of course, other factors could be responsible for this difference, because the type of replication and the factors present at the leading and lagging strands are different (38, 39).

One possible alternative we have explored elsewhere is that transcription generates single-strand regions and affects recombination bias (40). We found that gene transcription does not affect the strand bias observed for oligo-mediated recombination. Our results here strengthen that observation by showing that replication direction is critical to the strand bias.

There are eight possible mismatch pairs, and MutS protein has been shown to bind to each pair in vitro (41, 42). The four oligos used here generate four of those eight mismatches. Binding by MutS protein and repair by the MMR system have a similar hierarchical pattern for the eight mismatch pairs (42, 43). The pattern is G·T, A·C, A·A, G·G>T·T, T·C, A·G>C·C, where the C·C mismatch is very weakly bound and poorly corrected. Our studies support a similar pattern of repair efficiency with G·G>T·C>A·G>C·C. A 370-fold difference in repair exists between the well repaired G·G and the poorly repaired C·C mismatches (see lagging strand in Table 4). The same hierarchy and efficiency of correction were found for both the lagging and leading strands.

Because C·C mismatches are not recognized by MMR in other bacterial species (20, 44), this particular feature of oligo recombination might have the potential to create high recombination frequencies in other bacteria. For this reason and others, the ability to transfer the homologous recombination system to other bacterial species and even to eukaryotes would be very useful. Because ssDNA at the replication fork is bound by Ssb protein (45, 46), and Beta protein is important for the interaction between the oligo and the chromosome target, Beta may interact directly and specifically with Ssb (29). The λ Red system has been shown to work in Salmonella typhimurium (47–49), a species very closely related to E. coli, and it may also work in other related Gram-negative bacteria. However, for more distantly related bacteria, it may be necessary to provide Beta-like functions from phage endogenous to those bacteria. Exo- and Beta-like proteins have already been identified in other bacterial phages and even eukaryotic viruses such as HSV-1 (50, 51).

Mismatch repair functions are known to prevent DNA exchange between related species by blocking recombination between homologous but divergent sequences (homeologous recombination) (24). A difference between the role of mismatch repair in homeologous recombination and in replication is that in the former, MutS and MutL appear to be required, whereas MutH and UvrD helicase have less importance (24, 25, 52). It is believed that MutS and MutL bind mismatches generated during homeologous exchanges, and the binding itself aborts further recombination without causing repair (53, 54).

Our results suggest the inhibition of oligo-mediated recombination by MMR functions is more analogous to the correction of DNA replication errors than to the role of MMR functions during homeologous recombination. MMR functions tested, including UvrD and MutH, appear to remove mismatches generated during oligo recombination and replication (Table 2). Also, unlike homeologous recombination, Feinstein and Low (55) found that during E. coli conjugation between sequences with very few mismatches, the MMR system inhibited recombination and that, like oligo-mediated recombination, MutH and helicase are required. Perhaps, as we suggest for oligo recombination, the incoming strand transferred during normal conjugation is annealed at the replication fork. An alternative is that the recombination intermediates formed generate a new replication fork (39, 56, 57) and recruit the MMR complex. A recent discovery that ssDNA modification of murine embryonic stem cells is inhibited by MMR is consistent with our results and may indicate that in stem cells, the modification is also at the replication fork (58).

The in vivo recombination technologies we describe, due to their efficiency, accuracy, and simplicity, may replace classical in vitro genetic engineering techniques. The λ Red-mediated homologous recombination system, which we use as a genetic engineering tool, is particularly useful for modifying the genome of E. coli, as well as cloned genome segments from other organisms (29, 59, 60). This study creates new opportunities for genome modification and in vivo analyses of DNA mechanics. Using synthetic oligos, recombinants with the chromosome and episomes can occur at such high efficiencies that selection is not required. Oligos that create C·C mispairs are recombined into the DNA of cells at efficiencies approaching 25% among the survivors of electroporation. In other words, it might be possible to generate a C replacement of G anywhere in the chromosome at these high efficiencies if the change is not toxic to the cell. Recombination levels approaching 25% were also found for any oligo-generated mismatch in strains defective for the MMR functions tested. This advance in technology allows efficient genetic modifications and may permit nucleotide analogs and adducts to be incorporated directly into the chromosome for in vivo biochemical studies.

Watch the video: Homologous Recombination (August 2022).