Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Oligo- and dsDNA-mediated genome editing using a tetA dual selection system in Escherichia coli

  • Young Shin Ryu ,

    Contributed equally to this work with: Young Shin Ryu, Sathesh-Prabu Chandran

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Resources, Writing – original draft, Writing – review & editing

    ‡ Co-first authors

    Affiliation School of Energy and Chemical Engineering, Ulsan National Institute of Science and Technology (UNIST), Ulsan, Republic of Korea

  • Sathesh-Prabu Chandran ,

    Contributed equally to this work with: Young Shin Ryu, Sathesh-Prabu Chandran

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Resources, Writing – original draft, Writing – review & editing

    ‡ Co-first authors

    Affiliation School of Energy and Chemical Engineering, Ulsan National Institute of Science and Technology (UNIST), Ulsan, Republic of Korea

  • Kyungchul Kim,

    Roles Formal analysis, Resources

    Affiliation School of Life Sciences, Ulsan National Institute of Science and Technology (UNIST), Ulsan, Republic of Korea

  • Sung Kuk Lee

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Resources, Supervision, Validation, Writing – original draft, Writing – review & editing

    Affiliations School of Energy and Chemical Engineering, Ulsan National Institute of Science and Technology (UNIST), Ulsan, Republic of Korea, School of Life Sciences, Ulsan National Institute of Science and Technology (UNIST), Ulsan, Republic of Korea

Oligo- and dsDNA-mediated genome editing using a tetA dual selection system in Escherichia coli

  • Young Shin Ryu, 
  • Sathesh-Prabu Chandran, 
  • Kyungchul Kim, 
  • Sung Kuk Lee


The ability to precisely and seamlessly modify a target genome is needed for metabolic engineering and synthetic biology techniques aimed at creating potent biosystems. Herein, we report on a promising method in Escherichia coli that relies on the insertion of an optimized tetA dual selection cassette followed by replacement of the same cassette with short, single-stranded DNA (oligos) or long, double-stranded DNA and the isolation of recombinant strains by negative selection using NiCl2. This method could be rapidly and successfully used for genome engineering, including deletions, insertions, replacements, and point mutations, without inactivation of the methyl-directed mismatch repair (MMR) system and plasmid cloning. The method we describe here facilitates positive genome-edited recombinants with selection efficiencies ranging from 57 to 92%. Using our method, we increased lycopene production (3.4-fold) by replacing the ribosome binding site (RBS) of the rate-limiting gene (dxs) in the 1-deoxy-D-xylulose-5-phosphate (DXP) biosynthesis pathway with a strong RBS. Thus, this method could be used to achieve scarless, proficient, and targeted genome editing for engineering E. coli strains.


The bacterial genome has been previously manipulated via double-stranded (ds) DNA homologous recombination or short single-stranded (ss) DNA (oligo)-mediated genome engineering to modify bacterial strains. Several tools have been developed and used, such as multiplex automated genomic engineering (MAGE) and related approaches [1], zinc finger nucleases (ZFNs) [2], transcription activator-like effector nucleases (TALENs) [3], and clustered regularly interspaced short palindromic repeat (CRISPR)-associated Cas9 nucleases [4]. With recent advances in synthetic biology and metabolic engineering, genome modification has become important for the production of desired strains in basic and applied research [510]. However, each method has disadvantages compared with the others in terms of recombination and selection efficiency, insertion size, host genotype, and plasmid cloning. Therefore, there remains a demand for simple and efficient methods that can be easily used for multi-purpose genome engineering.

Conventional tools exploit separate positive- and negative-selection marker genes. Antibiotic-resistant marker genes such as neo, bla, cat, and tetA [1115] are used to select the recombinants, whereas SceI, sacB, rpsL, tolC, galK, and thyA [1618] have been used to select for positive-marker, gene-free clones; PCR-generated mutations in these negative-selection markers lead to the loss of their critical function, resulting in the selection of false-positive clones [1619]. A dual selection system using a single gene for both positive and negative selection is considered as a more promising strategy in selecting desired recombinants because of its double functionality and elimination of inactive-marker genes caused by non-specific mutations [15,20]. TetA exports tetracycline from bacterial cells, conferring resistance to the antibiotic tetracycline [2123], and pumps cadmium [24] and nickel [25] cations into the cell, resulting in cell death. However, it has been reported that tetA negative selection is much less effective in Escherichia coli than in other bacteria such as Salmonella [22].

Recently, a scarless genome modification method, the CRISPR-Cas9 system, has been developed using a double strand break for selection rather than antibiotic resistance and negative-selection markers. This system has limitations and complications such as high escape frequency, off-site effects, cellular toxicity of nucleases, plasmid cloning, and the requirement of a protospacer adjacent motif (PAM) sequence (5′-NGG-3ʹ) for specific targeting and expression of large Cas9 proteins [2628]. Moreover, it has been reported that point mutations in cas9 or elsewhere in the guide RNA plasmid result in cells that “escape” CRISPR-induced death [27]. Because of the high escape frequency, large DNA insertions are very difficult to obtain with the CRISPR-Cas9 system. More recently, Tas et al. [29] developed a genome modification method for E. coli using counter selection markers (either tetA or galK) in association with the λ-Red proteins SceI and RecA. In addition to high mutation efficiency, this method has disadvantages such as expression of multiple proteins and use of expensive chemicals and medium components to achieve maximum efficiency.

In this study, we developed a tetA-based dual selection system in E. coli for both short oligo- and long dsDNA-mediated genome editing. This method could be rapidly and successfully used for genome engineering without plasmid cloning. We demonstrated the efficacy of our system by optimizing metabolic flux through the 1-deoxy-D-xylulose-5-phosphate (DXP) biosynthesis pathway. We suggest that this system could achieve scarless, efficient targeted genome editing with a seamless screening process.

Materials and methods

Bacterial strains, plasmids, and oligomers

Wild-type E. coli MG1655 (MG) was used as the host strain for genome editing. The bacterial strains and plasmids as well as the oligomers used in this study are listed in Tables 1 and 2, respectively. Oligonucleotides were synthesized by Macrogen (Seoul, Korea).

Media and culture conditions

Bacterial strains were cultured in Luria-Bertani Miller broth (LB: 10 g of NaCl, 5 g of yeast extract, and 10 g of tryptone per liter) at 37°C, whereas cells containing temperature-sensitive plasmids were cultured at 30°C. For oligo transformation, E. coli cells were cultured in LB Lennox broth (LB-L: 5 g of NaCl, 5 g of yeast extract, and 10 g of tryptone per liter). The medium was supplemented with suitable antibiotics for plasmid maintenance and selection of positive-antibiotic markers at the following concentrations: chloramphenicol (Cm) at 30 μg/mL or tetracycline (Tet) at 10 μg/mL. For the nickel selection process, the cells were cultured in liquid minimal medium containing M9 salts (6.78 g of NaH2PO4, 3 g of KH2PO4, 0.5 g of NaCl, and 1 g of NH4Cl per liter), 0.2% glucose, and 50 μM NiCl2. The cells were incubated in a shaking incubator at 200 rpm at 37°C or 30°C.

Chromosomal tetA expression under different promoters

The tetA gene from pBBR1MCS3 was expressed on the E. coli chromosome under three different promoters: the tetA gene native promoter (PtetA) [32], the synthetic constitutive promoter CP25 (PCP25) [35], or the P3 promoter including a bicistronic ribosome binding site (RBS) termed P3-BCD2 (PP3BCD2) [36]. Construction of the dual selection cassettes containing tetA and the three different promoters is illustrated in S1 Fig, and the sequences of all three promoters are provided in S2 Fig. Briefly, tetA and PtetA were amplified using the primers PtetA-F and tetA–R; PCP25 was linked to tetA using the primers PCP25-F and tetA–R; and PP3BCD2 was linked to tetA using three primers (i.e., PP3BCD2-F1, PP3BCD2-F2, and tetA–R) by overlap extension PCR. The same cassettes were PCR-amplified with primers lacZ-H1P1 and tetA–R and then integrated into the lacZ region of the MG-Red strain expressing the λ-Red recombinase genes (i.e., gam, exo, and bet) on pSIM5 [31] to construct the dual selection cassette variant strains MG-PT, MG-PC, and MG-PP3. The pSIM5 plasmid contains the temperature-sensitive replication protein RepA101, the λ-Red recombinase genes under the PL promoter controlled by the cI857 temperature-sensitive repressor, and a chloramphenicol resistance marker gene. This plasmid can be cured simply by growing the cells at temperatures above 37°C and the λ-Red system can be easily switched on at 42°C and off at 32°C.

To identify the optimal NiCl2 and tetracycline concentrations for the dual selection process, the cell growth of the tetA-integrated strains (i.e., MG-PT, MG-PC, and MG-PP3) was monitored separately at different concentrations of NiCl2 (10−100 μM) and tetracycline (10−100 μg/mL). Cell growth was recorded by measuring absorption at 600 nm using a microplate reader (Infinite F200 PRO, Tecan) in a Corning 96-well, clear bottom plate with shaking (200 rpm, 37°C).

Genome modification

For oligo-mediated genome editing, the dual selection cassette PP3BCD2-tetA was integrated either between 838 and 839 bp without deletion or between 838 and 879 bp with a 40-bp deletion of lacZ to generate two nonfunctional-lacZ strains, MG-LacZ0 and MG-LacZ40, respectively. A 90-nt oligo (i.e., lacZ rescue 1) flanked with sequences homologous to the target regions was designed to rescue the disrupted lacZ of the strains MG-LacZ0 and MG-LacZ40. For dsDNA-mediated genome editing, a 1.5-kb region in lacZ was deleted by the insertion of the PP3BCD2-tetA, resulting in the nonfunctional-lacZ strain MG-LacZ1500. The deleted region with 300 bp of homologous sequences on either side was PCR-amplified from intact lacZ using the primers lacZ–resF and lacZ–resR. To rescue the disrupted lacZ in the strains MG-LacZ0, MG-LacZ40, and MG-LacZ1500, the strains were transformed with the respective oligo or 2.1-kb PCR products. Briefly, the strain MG carrying the plasmid pSIM5 and PP3BCD2-tetA on the chromosome was grown in 5 mL of fresh LB-L containing Cm and Tet, at 30°C. The λ-Red system was induced at an optical density (OD600) of approximately 0.5 by heating at 42°C for 15−20 min in a water bath as previously described [31]. Cells were immediately chilled on ice for at least 20 min and then rendered electrocompetent. Oligonucleotides were added to 50 μL of cells at a final concentration of 5 μM and electroporation was performed using a MicroPulser electroporator (Bio-Rad). Following electro-transformation, the cells were recovered by 3-h outgrowth at 30°C in 3 mL of LB. The cells were then diluted 1:100 and sub-cultured in liquid M9 medium containing 0.2% glucose and 50 μM of NiCl2 at 37°C for approximately 16 h. To measure rescue/selection efficiency, the sub-cultured cells were diluted to 10−6 prior to being spread onto LB plates containing X-gal/IPTG, and the efficiency was calculated by counting the blue colonies from the total number of colonies. If required, a second round of recombination was performed: following the first round of electro-transformation, the 3 mL of cells recovered by 3-h outgrowth at 30°C were induced at 42°C for 15−20 min and subsequently transformed with the oligos or dsDNA. Selection in liquid M9 medium with NiCl2 was then performed as described above.

Application to metabolic engineering

The PP3BCD2-tetA cassette was integrated into the 5′ untranslated region (5′-UTR) of dxs, which encodes one of the lycopene synthesis rate-limiting enzymes, in an E. coli strain expressing the λ-Red system on the plasmid pSIM5. A strong 5′-UTR (CTAGTTGGTAAGGAGTCTATAGTTG) was predicted using the UTR designer program [37]. The same cassette was then replaced with an oligo (i.e., UTR-dxs) containing a strong 5′-UTR flanked by the respective homologous sequences of dxs. Short ssDNA (oligos) transformation was performed twice and, after 3-h outgrowth, nickel selection was performed to obtain an engineered dxs strain, named MG-eDXS. The plasmid pINZ-1-LYC04, which contains heterologous genes required for the final steps of lycopene production [33,38], was transformed into MG and MG-eDXS resulting in the strains MG-Lyco and MG-eDXS-Lyco, respectively. For lycopene production, cells (i.e., MG-Lyco and MG-eDXS-Lyco) were cultured separately in 250-mL flasks containing 50 mL of LB broth at 30°C for 48 h. Lycopene was extracted with acetone and quantified by measuring absorbance (OD470). Lycopene content was calculated according to a standard curve as previously described [39]. To analyze DXS expression levels in MG-eDXS by Western blotting, a FLAG-tag sequence was inserted immediately after dxs in the strains MG and MG-eDXS resulting in the strains MG-DXS-FLAG and MG-eDXS-FLAG, respectively. Western blotting band intensity was measured using the ImageJ program (National Institutes of Health, USA). In addition, the FLAG-tag sequence was replaced in MG-DXS-FLAG and MG-eDXS-FLAG with gfp, which encodes green fluorescent protein (GFP), using our dsDNA homologous recombination genome editing method. This allowed determining dxs expression by measuring GFP intensity in the resultant variant strains MG-DXS-GFP and MG-eDXS-GFP, respectively. The genetic variants were confirmed by PCR and DNA sequencing.


A new strategy for genome editing

The scheme of the proposed dual selection genome editing strategy is illustrated in Fig 1. Step I: the plasmid pSIM5 carrying the λ-Red system was transformed into the strain MG and the transformants were selected on an LB plate containing Cm (30 μg/mL) at 30°C. Step II: the dual selection cassette PP3BCD2-tetA flanked by 50-nt homologous sequences of the target region was PCR-amplified and transformed into an E. coli strain expressing the λ-Red system. Recombinant cells possessing the tetA dual selection cassette on the chromosome were selected on an LB plate containing Tet (10 μg/mL) at 30°C. Red recombineering functions were induced in cells carrying the plasmid pSIM5 by heating at 42°C for 15 min and thus made electrocompetent. Step III: ss-oligos or dsDNAs with homologous sequences were transformed into cells containing PP3BCD2-tetA on the target region of the chromosome and expressing the λ-Red system. The dual selection cassette was replaced by either oligo-mediated recombination or dsDNA homologous recombination. Step IV: the genome-modified recombinants were enriched by culturing the cells in M9 liquid medium containing NiCl2 (50 μM) at 37°C. The plasmid pSIM5 was simultaneously cured during this step. The cultured cells were streaked onto LB plates and the desired recombinants were confirmed by colony PCR and DNA sequencing. To obtain improved selection efficiency, additional selection can be performed to select non-edited cells containing tetA using replica plating of colonies from LB to both LB and LB-Tet plates. Colonies that grow on LB but fail to grow on LB-Tet are the desired mutants.

Fig 1. Schematic diagram of oligo- and dsDNA-mediated genome editing using a tetA dual selection system in Escherichia coli.

Step I: Introduction of the λ-Red system into wild-type E. coli. Step II: Integration of the tetA dual selection cassette into the chromosomal target site. Step III: Genome editing by introduction of an ss-oligomer or dsDNA. Step IV: Selection of the genome-modified cells on NiCl2 under optimum selection conditions. A detailed description of this method is provided in the text.

Effective expression of chromosomal tetA for negative selection

Although the tetA negative-selection system is successful in Salmonella, it is much less effective in E. coli strains [22]. We hypothesized that TetA expression level is not sufficient for negative selection in E. coli. Thus, the relationship between the expression level of chromosomally encoded TetA and negative selection was tested by analyzing the growth of cells expressing tetA under different promoters (i.e., PtetA, PCP25, or PP3BCD2) at various concentrations of NiCl2 (up to 100 μM; Fig 2). The expression level of TetA under the three promoters was analyzed by measuring cell growth in the presence of high tetracycline concentration. Maximum growth was exhibited by PP3BCD2-tetA even at the highest concentration of tetracycline (100 μg/mL), indicating that tetA expression was sufficiently high to confer tetracycline resistance. All variants and wild-type strains were non-sensitive at 10 μM [Fig 2B(f)] and sensitive at 100 μM [Fig 2B(h)] NiCl2. Consistent with the tetracycline resistance results, when tetA was expressed under a known strong promoter, PP3BCD2, cell growth was significantly inhibited in the presence of NiCl2. At 50 μM NiCl2, wild-type cells could grow but cells expressing tetA under PP3BCD2 could not, indicating that negative selection is relatively effective at this concentration. The results indicate that both promoters (i.e., PtetA and PCP25) were not strong enough to express tetA and to confer strong nickel sensitivity (Fig 2). In contrast, the PP3BCD2-tetA system could be used for tetA dual selection in E. coli.

Fig 2. Promoter selection and optimization of NiCl2 and tetracycline concentrations for dual selection.

tetA was integrated into the strain MG (◇) with three different promoters (i.e., PtetA, PCP25, and PP3BCD2) and the resulting variant strains MG-PT (□), MG-PC (Δ), and MG-PP3 (×) were grown in M9 medium containing different concentrations of tetracycline (A) [0 (a), 10 (b), 50 (c), and 100 (d) μg/mL] and NiCl2 (B) [0 (e), 10 (f), 50 (g), and 100 (h) μM] to identify the strongest promoter and the optimal concentrations of Tet and NiCl2 for the dual selection process. PP3BCD2 was strong enough to express the TetA protein and confer strong nickel sensitivity at 50 μM NiCl2 and tetracycline resistance at the highest concentration (100 μg/mL), thereby allowing successful selection of the target-edited mutants and elimination of target-unedited cells. Data represent the mean value of three independent experiments and error bars represent standard deviation.

Selection efficiency of the tetA-based genome editing method

The selection efficiency of the proposed genome editing method was determined using the lacZ region to perform common genetic modifications. The nonfunctional-lacZ strains (i.e., MG-LacZ0 and MG-LacZ40) were rescued by deleting the integrated dual selection cassette with oligo lacZ rescue 1. In the case of MG-LacZ0 rescue, selection efficiency for the functional lacZ following two recombination events was as high as 92% with nickel selection (Fig 3A), whereas, approximately 4% of the cells were identified as recombinants without nickel selection. This result indicated that, following oligo transformation, subculturing the transformants in liquid M9 salt medium containing NiCl2 increased the selection efficiency. The same trend was also found for MG-LacZ40 rescue. Moreover, no colonies with the desired insertions were obtained by selection in liquid M9 salt medium without NiCl2. The selection efficiency for functional lacZ rescue in the strain MG-LacZ40 following two recombination events approached 79% (Fig 3B). Approximately, 30- and 80-fold increase in selection efficiency was achieved in MG-LacZ0 or MG-LacZ40, respectively, by subculturing the cells in liquid M9 salt medium containing NiCl2, compared to subculturing in M9 salt medium alone. MG-LacZ0 showed 13% higher efficiency than MG-LacZ40 did because it contains a homologous region for recombination longer than that found in MG-LacZ40. Additionally, there was no significant difference in selection efficiency when two or more recombination events were carried out (data not shown). Subsequent rounds of genome modification are not obligatory and, moreover, considerable efficiency was obtained following only one recombination event.

Fig 3. Genome editing using ssDNA oligos and linear dsDNA.

The lacZ segment was targeted in our genome editing method. The lacZ region was disrupted by the integration of the dual selection cassette PP3BCD2-tetA, which was rescued with either an oligo, [MG-LacZ0 (A) or MG-LacZ40 (B)] or linear dsDNA [(MG-LacZ1500 (C)]. Filled and unfilled bars represent the selection efficiency obtained with and without nickel selection, respectively. Numbers above the bar indicate the number of blue colonies (numerator) and total number of colonies (denominator) obtained by spreading the transformants on an LB plate containing X-gal/IPTG. (a), (b), and (c) refer to the editing steps followed to achieve target genome editing. Red and green in the oligos sequences represent the regions homologous to lacZ. Blue sequences represent the inserted bases. Data represent the mean value of three independent experiments and error bars represent standard deviation.

The efficiency of our system for dsDNA-mediated recombination was also evaluated by rescuing the nonfunctional-lacZ strain, MG-LacZ1500, with a PCR-amplified dsDNA fragment (2.1 kb). The efficiency was approximately 57%, similar to that of the oligo-mediated method (Fig 3C). Furthermore, no colonies with the desired insertions were obtained by selection in liquid M9 salt medium without NiCl2, indicating that tetA negative selection is very effective in E. coli. The results suggest that our system is also suitable for long, linear dsDNA-mediated genome editing even in cases of very low integration efficiency. The widely known constraints of oligo-mediated insertions, including size (≤ 120 nt) [1], can be circumvented by our dsDNA-mediated recombination method. We successfully achieved linear dsDNA-mediated genome insertion of a lacZ fragment at least 1.5 kb in size with flanking homology regions of at least 300 bp.

Application to metabolic engineering

To demonstrate the applicability of our system, we optimized the metabolic flux through the DXP biosynthesis pathway to overproduce lycopene in E. coli MG1655 (MG). The 5′-UTR of dxs, the gene encoding the rate-limiting enzyme in lycopene biosynthesis, was replaced with a strong 5′-UTR to enhance the translation efficiency. The engineered strain MG-eDXS-Lyco produced intense red pigmentation caused by increased lycopene production [Fig 4A(b)] compared to MG-Lyco [Fig 4A(b)]. Lycopene production was 3.4-fold in MG-eDXS-Lyco compared to that in MG-Lyco (Fig 4B). To confirm the increase in dxs expression, a FLAG-tag sequence was inserted immediately after dxs, and the expression of Dxs-FLAG was analyzed by SDS-PAGE [Fig 4C(a)] and Western blotting [Fig 4C(b)]. Analysis of the relative Western blotting band intensity of MG-eDXS-FLAG and MG-DXS-FLAG showed that dxs expression level was highly increased (approximately 4-fold) in MG-eDXS-FLAG compared to MG-DXS-FLAG. Gene expression level can be measured by detecting the fluorescence intensity of gfp fused to the target gene. The gfp gene was integrated downstream of dxs using our method, resulting in the strains MG-DXS-GFP and MG-eDXS-GFP, which was confirmed by PCR and DNA sequencing (S3 Fig). However, GFP expression could not be detected, possibly due to low expression levels on the chromosome.

Fig 4. Application in metabolic engineering.

Lycopene production was enhanced by increasing the translation efficiency of dxs in the lycopene synthesis pathway. The MG-eDXS-Lyco cell pellet [A(b)] showed intense red pigmentation compared to MG-Lyco [A(a)]. MG-eDXS-Lyco produced 3.4-fold lycopene amount than MG-Lyco did (B). dxs expression in MG-DXS-FLAG and MG-eDXS-FLAG was analyzed by SDS-PAGE [C(a)] and Western blotting [C(b)].


Successful selection of genome-edited recombinants requires high editing efficiency or high-throughput screening. One approach for increasing recombineering efficiency is λ-Red-assisted oligonucleotide-mediated recombination [40]. However, replacement with oligos containing more than five mismatched bases and the deletion of entire genes result in relatively low frequency of genomic modifications, hindering the selection of particular genetic variants from a large number of unedited colonies [41]. In addition, this method cannot be used for large dsDNA recombination with low efficiency because it does not include selection strategies. These limitations were overcome in our system by combining a tetA-based dual selection system with dsDNA homologous recombination or oligo-mediated recombination. Our system exhibits high selection efficiency, i.e., 57% and 92% for dsDNA and oligo-mediated recombineering, respectively.

Additionally, our method overcomes the problems of the popular tetA-sacB genome editing method, including the size of the tetA-sacB cassette, inefficiency of the counter-selection marker due to mutation in sacB, and the need for complex and expensive growth medium for counter-selection [15]. Mizoguchi et al. [42] reported that approximately 76% of the transformants presented false-positive phenotype when using the negative-selection marker sacB and the positive-selection marker cat due to mutations in the negative-selection marker gene sacB. The use of independent selection markers for positive and negative selections significantly increases the chances of the emergence of false positives due to the mutations in the negative selection marker. This problem has been addressed by the use of a single selection marker such as tetA as a dual selection marker because any mutation that affects the function of the dual selection marker protein will likely affect the both positive and negative selections [43,44], thus, significantly reducing the complications of false positives. In the tetA dual selection system, non-functional tetA mutants could be selected out during the first positive selection because the mutants cannot survive on an LB-Tet plate. The recently developed TetA/galK-SceI method takes 1−2 weeks from start to verified end-product [29]. The use of galK has disadvantages as the recipient host needs to be galK, and counter-selection is performed using a medium with either an expensive galactose analogue, 2-deoxy-galactose, or low-cost NiCl2 [29]. Unlike other methods [1,15,27,29], our method can be performed rapidly at low cost in liquid minimal salt medium containing NiCl2. Our method accomplishes successful recombination in a short time frame (7−8 days) compared with other similar scarless genome editing methods such as CRISPR and SecI (S1 Table). Comparisons of our method with other popular scarless genome editing methods are detailed in Table 3.

The expression level of tetA, which confers metal sensitivity, is a key factor for employing the tetA dual selection system in genome editing applications [25,29,45]. Consistent with a previous study [22], nickel sensitivity for negative selection was not observed in cells harboring tetA on the chromosome under the control of either the native promoter PtetA or the synthetic constitutive promoter PCP25, which is known to be suitable for ensuring strong expression in E. coli [35]. However, when tetA was expressed under PP3BCD2, a known strong promoter with a double RBS [36], nickel had toxic effects on cell growth in liquid M9 salt medium supplemented with 50 μM NiCl2. These findings indicate that tetA expression under PP3BCD2 was higher than under the other two promoters tested, and eventually resulted in nickel sensitivity, which might be due to sufficient expression of TetA for negative selection against nickel. We believe that the method proposed in this study could solve the problems linked to the tetA system and that it can be used effectively in E. coli for scarless and seamless genome editing.

Since dsDNA homologous recombination-mediated genome editing is much less efficient than the oligo-mediated method [46], target-edited cells were not obtained without the nickel selection step when a large dsDNA (2.1 kb) was inserted by homologous recombination. Following nickel negative selection, approximately 57% of the transformants were target-edited mutants, indicating that, although recombination efficiency was low, the genetic recombinants were enriched and the unedited cells were eliminated by nickel negative selection. In this study, we obtained high selection efficiency for both oligo-mediated and dsDNA homologous recombination under the optimized selection conditions. We suggest that this feature could be effectively employed for various genome-editing projects when selection or screening processes are unavailable.

Oligo-mediated genome engineering has high editing efficiency (maximum theoretical efficiency per recombination cycle is 25%) [1,41], producing many genome variants. Unlike dsDNA homologous recombination, this method does not include a selection process, resulting in difficulties in identifying specific recombinants, namely gene deletion and replacement of oligos presenting long mismatching sequences with low editing efficiency (< 0.1%). Low efficiency makes it difficult to select target recombinants from unedited cells. This problem can be solved by increasing the selection efficiency using the tetA dual selection system employed in our method.


Here, we developed a reliable and versatile genome editing method by combining the novel application of oligo- or dsDNA-mediated genome editing and a tetA dual selection system that offers scarless genome modifications in E. coli. This method has additional advantages over other scarless genome editing tools. First, there is no need for plasmid cloning, thereby making it a simple, time-saving, and less labor-intensive method. Second, screening of recombinants by tetA dual selection renders the method very efficient as it reduces the number of false-positive mutants. Altogether, the method described herein is more robust than others and enables highly efficient manipulation of the bacterial genome, including the rewiring of metabolic pathways as well as the construction of new microbial cell factories for applied research and industrial applications.

Supporting information

S1 Fig. Construction of tetA dual selection cassettes with three different promoters.

tetA amplified from pBBR1MCS3 was linked to three different promoters: the tetA gene native promoter (PtetA) [32], synthetic constitutive promoter CP25 (PCP25) [35], or the P3 promoter including a bicistronic ribosome binding site termed P3-BCD2 (PP3BCD2) [36], to construct the three different dual selection cassettes PtetA-TetA, PCP25-TetA, and PP3BCD2-TetA, respectively. These cassettes were individually transformed into MG-Red to construct dual selection cassette-variant strains, MG-PT, MG-PC, and MG-PP3. Detailed information is provided in the text.


S2 Fig. Nucleotide sequences of the promoters used for chromosomal tetA expression in the lacZ region.

(A) PtetA: native promoter of tetA [32]; (B) PCP25: synthetic constitutive promoter CP25 [35]; (C) PP3BCD2: synthetic constitutive promoter P3 including bicistronic ribosome binding site [36]; P1 and P4: priming sequences [47]; H1 and H2: upstream and downstream homologous sequences; −35 and −10: promoter regions; +1: transcription start site.


S3 Fig. Gene insertion using dsDNA-homologous recombination.

(A) Agarose gel electrophoresis of the PCR products following the insertion of gfp downstream of dxs (B) Steps and primers (dxs-seqF2 and dxs-seqR2) used to confirm recombination. The size of specific PCR products is given in parentheses.


S1 Table. Timeline of the workflow for the tetA dual selection.



  1. 1. Wang HH, Isaacs FJ, Carr PA, Sun ZZ, Xu G, Forest CR, et al. Programming cells by multiplex genome engineering and accelerated evolution. Nature. 2009;460: 894–898. pmid:19633652
  2. 2. Carroll D. Genome engineering with zinc-finger nucleases. Genetics. 2011;188: 773–782. pmid:21828278
  3. 3. Mahfouz MM, Li L, Shamimuzzaman M, Wibowo A, Fang X, Zhu J-K. De novo-engineered transcription activator-like effector (TALE) hybrid nuclease with novel DNA binding specificity creates double-strand breaks. Proc Natl Acad Sci U S A. 2011;108: 2623–2628. pmid:21262818
  4. 4. Cong L, Ran FA, Cox D, Lin S, Barretto R, Habib N, et al. Multiplex genome engineering using CRISPR/Cas systems. Science. 2013;339: 819–823. pmid:23287718
  5. 5. Jeong J, Cho N, Jung D, Bang D. Genome-scale genetic engineering in Escherichia coli. Biotechnol Adv. 2013;31: 804–810. pmid:23624241
  6. 6. Kim J, Webb AM, Kershner JP, Blaskowski S, Copley SD. A versatile and highly efficient method for scarless genome editing in Escherichia coli and Salmonella enterica. BMC Biotechnol. 2014;14: 84. pmid:25255806
  7. 7. Pyne ME, Moo-Young M, Chung DA, Chou CP. Coupling the CRISPR/Cas9 system to lambda Red recombineering enables simplified chromosomal gene replacement in Escherichia coli. Appl Environ Microbiol. 2015;81: 5103–5114. pmid:26002895
  8. 8. Kim SM, Choi BY, Ryu YS, Jung SH, Park JM, Kim GH, et al. Simultaneous utilization of glucose and xylose via novel mechanisms in engineered Escherichia coli. Metab Eng. 2015;30: 141–148. pmid:26045332
  9. 9. Vinuselvi P, Lee SK. Engineered Escherichia coli capable of co-utilization of cellobiose and xylose. Enzyme Microb Technol. 2012;50: 1–4. pmid:22133432
  10. 10. Vinuselvi P, Park JM, Lee JM, Oh K, Ghim CM, Lee SK. Engineering microorganisms for biofuel production. Biofuels. 2011;2: 153–166.
  11. 11. Zhang Y, Buchholz F, Muyrers JP, Stewart AF. A new logic for DNA engineering using recombination in Escherichia coli. Nat Genet. 1998;20: 123–128. pmid:9771703
  12. 12. Liang R, Liu J. Scarless and sequential gene modification in Pseudomonas using PCR product flanked by short homology regions. BMC Microbiol. 2010;10: 209. pmid:20682065
  13. 13. Ellis HM, Yu D, DiTizio T, Court DL. High efficiency mutagenesis, repair, and engineering of chromosomal DNA using single-stranded oligonucleotides. Proc Natl Acad Sci U S A. 2001;98: 6742–6746. pmid:11381128
  14. 14. Thomason LC, Sawitzke JA, Li X, Costantino N, Court DL. Recombineering: genetic engineering in bacteria using homologous recombination. Curr Protoc Mol Biol. 2007;78: V:1.16.1–1.16.24.
  15. 15. Li X-T, Thomason LC, Sawitzke JA, Costantino N, Court DL. Positive and negative selection using the tetA-sacB cassette: recombineering and P1 transduction in Escherichia coli. Nucleic Acids Res. 2013;41: e204. pmid:24203710
  16. 16. Wong QNY, Ng VCW, Lin MCM, Kung H, Chan D, Huang JD. Efficient and seamless DNA recombineering using a thymidylate synthase A selection system in Escherichia coli. Nucleic Acids Res. 2005;33: e59. pmid:15800210
  17. 17. DeVito JA. Recombineering with tolC as a selectable/counter-selectable marker: remodeling the rRNA operons of Escherichia coli. Nucleic Acids Res. 2008; 36: e4. pmid:18084036
  18. 18. Heermann R, Zeppenfeld T, Jung K. Simple generation of site-directed point mutations in the Escherichia coli chromosome using Red®/ET® Recombination. Microb Cell Fact. 2008;7: 14. pmid:18435843
  19. 19. Warming S, Costantino N, Court DL, Jenkins NA, Copeland NG. Simple and highly efficient BAC recombineering using galK selection. Nucleic Acids Res. 2005;33: e36. pmid:15731329
  20. 20. Madyagol M, Al-Alami H, Levarski Z, Drahovská H, Turňa J, Stuchlík S. Gene replacement techniques for Escherichia coli genome modification. Folia Microbiol (Praha). 2011;56: 253–263.
  21. 21. Schnappinger D, Hillen W. Tetracyclines: antibiotic action, uptake, and resistance mechanisms. Arch Microbiol. 1996;165: 359–369. pmid:8661929
  22. 22. Bochner BR, Huang HC, Schieven GL, Ames BN. Positive selection for loss of tetracycline resistance. J Bacteriol. 1980;143: 926–933. pmid:6259126
  23. 23. Maloy SR, Nunn WD. Selection for loss of tetracycline resistance by Escherichia coli. J Bacteriol. 1981;145: 1110–1111. pmid:7007341
  24. 24. Griffith JK, Buckingham JM, Hanners JL, Hildebrand CE, Walters RA. Plasmid-conferred tetracycline resistance confers collateral cadmium sensitivity to E. coli cells. Plasmid. 1982;8: 86–88. pmid:6755512
  25. 25. Podolsky T, Fong ST, Lee BTO. Direct selection of tetracycline-sensitive Escherichia coli cells using nickel salts. Plasmid. 1996;36: 112–115. pmid:8954882
  26. 26. Esvelt KM, Wang HH. Genome scale engineering for systems and synthetic biology. Mol Syst Biol. 2013;9: article number 641. pmid:23340847
  27. 27. Reisch CR, Prather KLJ. The no-SCAR (Scarless Cas9 Assisted Recombineering) system for genome editing in Escherichia coli. Sci Rep. 2015;5: 15096. pmid:26463009
  28. 28. Gupta RM, Musunuru K. Expanding the genetic editing tool kit: ZFNs, TALENs, and CRISPR-Cas9. J Clin Invest. 2014;124: 4154–4161. pmid:25271723
  29. 29. Tas H, Nguyen CT, Patel R, Kim NH, Kuhlman TE, Warming S, et al. An integrated system for precise genome modification in Escherichia coli. PLoS One. 2015;10: e0136963. pmid:26332675
  30. 30. Blattner F, Plunkett G 3rd, Bloch C, Perna N, Burland V, Riley M, et al. The complete genome sequence of Escherichia coli K-12. Science. 1997;277: 1453–1462. pmid:9278503
  31. 31. Datta S, Costantino N, Court DL. A set of recombineering plasmids for gram-negative bacteria. Gene. 2006;379: 109–115. pmid:16750601
  32. 32. Kovach ME, Elzer PH, Hill DS, Robertson GT, Farris MA, Roop RM, et al. Four new derivatives of the broad-host-range cloning vector pBBR1MCS, carrying different antibiotic-resistance cassettes. Gene. 1995;166: 175–176. pmid:8529885
  33. 33. Ryu YS, Biswas RK, Shin K, Parisutham V, Kim SM, Lee SK. A simple and effective method for construction of Escherichia coli strains proficient for genome engineering. PLoS One. 2014;9: e94266. pmid:24747264
  34. 34. Huston J, Levinson D, Mudgett-Hunter M, Tai M, Novotnyý J, Margolies M, et al. Protein engineering of antibody binding sites: recovery of specific activity in an anti-digoxin single-chain Fv analogue produced in Escherichia coli. Proc Natl Acad Sci U S A. 1988;85: 5879–5883. pmid:3045807
  35. 35. Jensen PR, Hammer K. The sequence of spacers between the consensus sequences modulates the strength of prokaryotic promoters. Appl Environ Microbiol. 1998;64: 82–87. pmid:9435063
  36. 36. Mutalik VK, Guimaraes JC, Cambray G, Lam C, Christoffersen MJ, Mai QA, et al. Precise and reliable gene expression via standard transcription and translation initiation elements. Nat Methods. 2013;10: 354–360. pmid:23474465
  37. 37. Seo SW, Yang JS, Kim I, Yang J, Min BE, Kim S, et al. Predictive design of mRNA translation initiation region to control prokaryotic translation efficiency. Metab Eng. 2013;15: 67–174. pmid:23164579
  38. 38. Kang MJ, Lee YM, Yoon SH, Kim JH, Ock SW, Jung KH, et al. Identification of genes affecting lycopene accumulation in Escherichia coli using a shot‐gun method. Biotechnol Bioeng. 2005;91: 636–642. pmid:15898075
  39. 39. Kim SW, Keasling JD. Metabolic engineering of the nonmevalonate isopentenyl diphosphate synthesis pathway in Escherichia coli enhances lycopene production. Biotechnol Bioeng. 2001;72: 408–415. pmid:11180061
  40. 40. Murphy KC. Use of bacteriophage lambda recombination functions to promote gene replacement in Escherichia coli. J Bacteriol. 1998;180: 2063–2071. pmid:9555887
  41. 41. Pines G, Freed EF, Winkler JD, Gill RT. Bacterial recombineering: genome engineering via phage-based homologous recombination. ACS Synth Biol. 2015;4: 1176–1185. pmid:25856528
  42. 42. Mizoguchi H, Tanaka-Masuda K, Mori H. A simple method for multiple modification of the Escherichia coli K-12 chromosome. Biosci Biotechnol Biochem. 2007;71: 2905–2911. pmid:18071259
  43. 43. Muranaka N, Sharma V, Nomura Y, Yokobayashi Y. An efficient platform for genetic selection and screening of gene switches in Escherichia coli. Nucleic Acids Res. 2009;37: e39. pmid:19190095
  44. 44. Nomura Y, Yokobayashi Y. Dual selection of a genetic switch by a single selection marker. Biosystems. 2007;90: 115–120. pmid:16942834
  45. 45. Ng CY, Farasat I, Maranas CD, Salis HM. Rational design of a synthetic Entner-Doudoroff pathway for improved and controllable NADPH regeneration. Metab Eng. 2015;29: 86–96. pmid:25769287
  46. 46. Sawitzke JA, Thomason LC, Costantino N, Bubunenko M, Datta S, Court DL. Recombineering: in vivo genetic engineering in E. coli, S. enterica, and beyond. Methods Enzymol. 2007;421: 171–199. pmid:17352923
  47. 47. Datsenko KA, Wanner BL. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci U S A. 2000;97: 6640–6645. pmid:10829079