https://orcid.org/0000-0001-5447-738X
Figures
Abstract
Seamless DNA vectors derived from bacterial plasmids are devoid of bacterial genetic elements and represent attractive alternatives for biomedical applications including DNA vaccines. Larger scale production of seamless vectors employs engineered Escherichia coli strains in order to enable tightly regulated expression of site-specific DNA recombinases which precisely delete unwanted sequences from bacterial plasmids. As a novel component of a developing lambda integrase genome editing platform, we describe here strain MG1655-ISC as a means to easily produce different scales of seamless vectors, ranging in size from a few hundred base pairs to more than ten kilo base pairs. Since we employed an engineered lambda integrase that is able to efficiently recombine pairs of DNA crossover sites that differ in sequence, the resulting seamless vectors will be useful for subsequent genome editing in higher eukaryotes to accommodate variations in target site sequences. Future inclusion of single cognate sites for other genome targeting systems could enable modularity. These features, together with the demonstrated simplicity of in vivo seamless vector production, add to their utility in the biomedical space.
Citation: Roy S, Peter S, Dröge P (2022) Versatile seamless DNA vector production in E. coli using enhanced phage lambda integrase. PLoS ONE 17(9): e0270173. https://doi.org/10.1371/journal.pone.0270173
Editor: Chen Ling, Fudan University, CHINA
Received: June 5, 2022; Accepted: September 7, 2022; Published: September 23, 2022
Copyright: © 2022 Roy et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: This work was supported through a grant from the National Research Foundation-Competitive Research Programme, Singapore to P.D. (NRF-CRP21-2018-0002). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: P.D. is a founder and stakeholder of LambdaGen Therapeutics Pte Ltd. S.R., S.P. and P.D. are listed as co-inventors on a related patent application.
Introduction
Seamless DNA vectors—also referred to as “minicircles or minimized vectors” [1, 2]—are derived from bacterial plasmids and devoid of bacterial genetic elements such as origins of replication and resistance markers. Seamless vectors are circular, covalently closed and usually negatively supercoiled DNA molecules. They are becoming increasingly attractive for biomedical applications such as cell line engineering, biologics production, gene/cell therapy and DNA vaccination. This attractiveness is due to their enhanced transgene expression, improved safety features, reduced gene silencing, and higher gene transfer efficiency when compared to parental plasmids [3, 4].
The most common means to produce seamless DNA vectors are site-specific DNA recombinases which utilize their cognate DNA sequences that flank unwanted bacterial genetic elements. Through a precise DNA strand cutting and pasting reaction at their respective cognate sites, these enzymes can splice out intervening DNA from the rest of the molecule. The recombination reaction using plasmids as substrate thus results in two circular DNA molecules: one that carries the unwanted bacterial elements and the other representing the desired seamless vector. The site-specific recombination reaction can be carried out either in vitro with purified enzymes for cost-effective small scale production or in vivo inside the bacterium Escherichia coli to achieve medium to large scale production; the latter is currently being employed in a few commercial settings. Various protocols are available to isolate and purify seamless vectors for downstream purposes at industrial scales [5].
A number of recombinases have been utilized for in vivo production of seamless vectors in E. coli and include the wild-type phage lambda integrase [6], the yeast recombinases Cre and FLP [7, 8], the wild-type phage ΦC31 integrase [9], and the ParA resolvase [10]. For in vivo seamless vector production, a technical challenge is the stringent control of the expression of the recombinase. Leaky expression during bacterial growth will result in premature loss of the seamless vector inside the bacterium and hence severely compromises yields. In order to achieve stringent recombinase expression control, several expression systems have been employed including temperature-sensitive lambda repressor cI857/pR [6] and the plasmid-based pBAD/araC arabinose system [9]. However, employing these systems reproducibly and at larger scales remains technically challenging due to low recombination efficiency resulting in less yield of seamless vector and contamination with parental plasmids in the final preparation of seamless vectors [11, 12].
In the present study, we present an engineered E. coli strain that can easily be used for multi-scale and multi-purpose seamless vector production by in vivo site-specific recombination catalyzed by an enhanced mutant lambda integrase, dubbed IntC3 [13]. Distinct advantages of this system are the ease of handling and, due to the employment of enhanced IntC3, the application of various pairs of attachment (att) recombination site derivatives as efficient recombination substrates to yield seamless vectors ranging in size between a few hundred base-pairs to > 10 kb. This flexibility in att site sequences considerably expands the scope of future downstream applications for seamless vectors including site-specific genome editing of higher eukaryotic cells [14, 15].
Materials and methods
Engineering of E. coli strain MG1655 carrying an inducible IntC3 expression cassette
The E. coli strain MG1655 was chosen as a base for generating a versatile seamless vector bacterial producer strain because it approximates K12 wild-type cells with minimal prior genetic changes [16]. We hypothesized that this feature provides us with higher chances of success for tightly regulated expression of IntC3 via the endogenous arabinose (ara) operon. MG1655 has been maintained as a laboratory strain with minimal genetic manipulation, having only been cured of the temperate bacteriophage lambda and F plasmid by means of ultraviolet light and acridine orange, respectively [17, 18].
A policistronic sequence of 2725 bp, termed ISC, composed of the IntC3, single chain integration host factor scIHF2 [19] and chloramphenicol resistance gene (CAT) was commercially synthesized (GeneScript) (S1 Fig). scIHF2 (single polypeptide chain IHF) is a modified version of wild-type IHF composed of the alpha subunit of IHF inserted into the beta subunit through a novel protein engineering approach. It exhibits similar properties to wild-type IHF in term of DNA binding and bending [19]. The chosen strategy for MG1655 engineering included the precise insertion of the ISC expression cassette into the ara operon of MG1655 immediately downstream of the promoter by using the start codon of the endogenous araB gene as start codon for INTC3 (Fig 1A). Two primers were designed to insert the construct at this locus using routine protocols of lambda red-mediated homologous recombination reactions [20]. The ATG start codon of the IntC3 gene was re-introduced in the forward primer (Fig 1B). PCR amplification of the ISC construct for electroporation into electroporation-competent MG1655 cells was performed with primers: INTC3_ARAB_FWD_HR:ACTCTCTACTGTTTCTCCATACCCGTTTTTTTGGATGGAGTGAAACGATGGGAAGAAGGCGAAGTCATGAGC and FRT_ARAB_REV_HR: GCCAAAGCTCGCACAGAATCACTGCCAAAATCGAGGCCAATTGCAATCGCTTATACAGTCGAAGTTCCTATA using Q5 High Fidelity DNA Polymerase (New England Biolabs). The thermal cycling parameters used for PCRs were as follows: initial denaturation at 98°C for 30 seconds, 35 cycles of denaturation at 98°C for 10 seconds, annealing at 65°C for 30 seconds and extension at 72°C for 2 minutes, and a final step of 72°C for 2 minutes. The resulting purified PCR product was used for subsequent electroporation into MG1655/pKD46 electroporation competent cells described below.
(A) Diagram of the ara operon with the inserted gene cassette composed of integrase variant C3 (INTC3), single chain integration host factor 2 (IHF), recombination sites for Flp recombinase (FRT) as direct repeats and chloramphenicol resistance cassette (CAT). The direction of transcription of the modified ara operon is indicated. (B) The red arrow demarcates the precise position of ISC insertion by homologous recombination between start and alanine codons of araB as verified by sequencing.
MG1655/pKD46 cells were made competent for plasmid transformation using standard protocols. Briefly, plasmid pKD46 [21] was transformed, cells plated on ampicillin selection media, and grown at 30°C overnight. A single colony of pKD46-transformed MG1655 was grown overnight in DYT media plus ampicillin (200μg/ml) at 30°C. The stationary culture was diluted 1:200 in fresh 200 ml DYT media and incubated at 30°C. When OD600 had reached 0.4, the culture was induced with 1.2% L-arabinose and incubated at 37°C for 1 hr at 180 rpm. Cells were immediately chilled in an ice water bath. The culture was left on ice for 20 minutes with occasional agitation. Subsequently, cells were kept chilled and distributed into 4 X 50 ml Falcon tubes on ice. Centrifugation was at 1000g for 10 to 20 minutes at 4°C. Once the supernatant became clear, cell pellets were suspended in 25 ml of double-deionized (DD) ice-cold water. Contents of 2 X 50 ml Falcon tubes were combined and centrifuged again under the same conditions. After the supernatant became clear, each pellet was suspended in 50ml ice cold water and centrifugation was repeated. The supernatant was discarded and the pellet suspended in 50 ml ice cold 10% glycerol (diluted in DD water). The centrifugation step was repeated and the pellet re-suspended in 0.5 ml 10% glycerol. The two re-suspended pellets were combined in 2 ml pre-chilled Eppendorf tubes and mixed. OD600 of a 1:100 diluted solution (10 μl suspended cells + 990 μl of 10% glycerol) was set 0.4 to 0.6. Competent cells were stored in aliquots of 70μl-100μl per tube after flash freezing in liquid nitrogen and stored at minus 80°C.
Electroporation of the ISC PCR-amplified construct into electroporation competent MG1655/pKD46 cells was performed with Gene Pulser (BioRad) as follows: 1 to 10 ng of the PCR product were added to 100 μl competent E. coli and electroporated in pre-set conditions (set 1 or 2). Cell recovery was at 37°C for 1 hr in DYT without antibiotics. The transformed cells were spread onto DYT media + 0.1% Glucose + 15 μg/ml chloramphenicol agar plates and grown at 30°C. Growth at 37°C will subsequently lead to the loss of pKD46 plasmid since it carries a temperature sensitive origin of replication. Resulting colonies were tested by colony PCR to confirm the left and right junctions of the inserted transgene cassette, as well as the presence of the entire cassette by genomic PCR (S2 Fig).
Primers for left junction PCR resulting in a 291 bp product: ARAC_FWD GTCTATAATCACGGCAGAAAAGTCC and INTC3_REV TCGCCTGTCTCTGCCTAATCC. Right Junction primers will yield a PCR product of 397 bp: CAT_FWD CGCAAGGCGACAAGGTGCT and ARAB_REV CCGCTTCCATTGACTCAATGTAGTC. Genomic PCR using ARAC_FWD and ARAB REV primers will generate a 2.85 kb product.
Colony PCR was performed as follows: One colony was diluted in 50 μl DYT media + 0.1% Glucose + 15 μg/ml chloramphenicol media and 2 μl used as colony PCR template. PCR was performed using GoTaq Flexi DNA polymerase (Promega) to amplify both the junctions in 25 μl reactions. The thermal cycling parameters used for PCRs were as follows: initial denaturation at 95°C for 5 minutes, 35 cycles of denaturation at 95°C for 1 minute, annealing at 56°C for 30 seconds and extension at 72°C for 1 minute, and a final step of 72°C for 5 minutes. PCR products were verified by sequencing. We dubbed our engineered strain MG1655-ISC.
Results
Growth of engineered MG1655 strains carrying integrase expression cassettes
Our MG1655-ISC strain retained at this stage the CAT gene as part of the transgene cassette. By including the FRT sequences for the yeast Flp recombinase (Fig 1), the option remained to remove the CAT expression cassette by transient Flp recombinase expression at a later stage. A critical parameter of any engineered E.coli strain for seamless vector production is the cell doubling time. We analyzed growth rates of MG1655-ISC and compared it with that of another engineered MG1655 strain that carries the same transgene expression cassette at the ara locus, except that scIHF2 [19] had been omitted. The latter strain was dubbed MG1655-IC and had been generated in parallel to MG1655-ISC following the same protocol.
An example of the growth rate analysis (S3 Fig) revealed that the presence of bothIntC3 and scIHF2 in the ara operon had no substantial effects on exponential growth rates, with cell doubling times in the typical range of 20 to 30 minutes. Furthermore, the presence of scIHF2 does not change the cell density at stationary phase.
Outline of multi-scale seamless vector production using MG1655-ISC
The general workflow for seamless vector production using strain MG1655-ISC is summarized in Fig 2. A plasmid with a standard bacterial backbone is flanked by two directly repeated lambda integrase recombination sequences, termed att1 and att2, and carries the desired DNA payload for seamless vector production. After transformation into MG1655-ISC, cells were grown with antibiotics until OD(600) reached 1.0. IntC3 and scIHF2 expression was induced by the addition of arabinose (1.5% final concentration), and cells were incubated for additional 70–90 minutes at 37°C.
See text for details. The catenated DNA rings generated by recombination inside E. coli are efficiently resolved into two monomeric DNA circles by topoisomerase IV. The DNA circles were purified from the E. coli strains and digested with an appropriate restriction enzyme to linearize the bacterial backbone and unrecombined substrate plasmid. This was followed by exonuclease treatment to digest the linear and nicked circular DNA products. Hence, only supercoiled seamless vector will be retained.
Induction of IntC3 leads to recombination between att1 and att2 which generates a dimeric DNA catenane consisting of one DNA ring that carried the bacterial backbone and a second DNA ring that carried the DNA payload plus one copy of a hybrid att sequence. The catenated DNA rings can be efficiently unlinked inside E.coli by endogenous type 2 topoisomerases.
Following induction of transgene expression, episomal DNA is purified from lysed MG1655-ISC cells by standard procedures, and the circular bacterial DNA is linearized by restriction digestion. The linearized DNA and contaminating nicked DNA molecules can be effectively degraded with phage T5 exonuclease. The remaining DNA is the intact, covalently closed, supercoiled seamless vector.
Mini-seamless vector production using variants of attL and attB sites in MG1655-ISC
To demonstrate broad utility of MG1655-ISC for seamless vector production, we first transformed the 6.3 kb substrate plasmid pattPhae2 (attL) (Fig 3A). This recombination substrate carried a 21 bp attB variant and a 121 bp attL variant recombination sequence in direct repeat orientation separated by about 530 bp. Recombination by IntC3 will result in two circular DNA molecules: a small supercoiled 530 bp mini-seamless vector and 5.8 kb supercoiled DNA that carries the bacterial backbone and other sequences.
(A) Map of recombination substrate pattPhae2 (attL). The attB and attL (PT2) sequences are derivatives from the corresponding wild-type attB and attL and were placed about 500 bp apart as direct repeats. Recombination deletes the DNA segment flanked by the two att sites. (B) A suspension culture (10ml) was induced for integrase expression by arabinose and further incubation for 70 minutes. Plasmid DNA was purified and analysed by agarose gel electrophoresis. Lane L: marker ladder; Lane 1: substrate vector attPhae2 (attL) undigested; Lane 2: substrate vector attPhae2(attL) after induction, undigested; Lane 3: substrate vector NdeI digested; Lane 4: substrate vector ScaI digested; Lane 5: purified plasmid DNA after arabinose induction was digested with NdeI and T5 exonuclease; 3μl of 100μl total sample loaded; Lane 6: same as lane 5, 6 μl loaded; Lane 7: Purified plasmid DNA after arabinose induction digested with ScaI. Note that the supercoiled and linearized seamless vectors run at about the same position.
MG1655-ISC cells transformed with pattPhae2 (attL) were induced by arabinose for 70 minutes and harvested. Episomal DNA was isolated and analyzed by agarose gel electrophoresis. The results (Fig 3B) revealed that compared to substrate DNA (lane 1), induction of IntC3 expression produced recombination products which were decatenated in vivo by endogenous topoisomerases, i.e., the two recombination product DNA rings were no longer physically (topologically) linked (lane 2; diagrammed in Fig 2). The released supercoiled mini-seamless vector (lane 2; arrow) migrates far ahead from the rest of the isolated DNA due to its small size. Restriction digestion of substrate DNA and recombination products by ScaI confirmed that the vast majority of the substrate had been recombined inside MG1655-ISC cells after arabinose induction (lanes 4 and 7, respectively). Restriction digestion of DNA isolated from induced cells by NdeI, which cleaves only the bacterial backbone segment, in the presence of T5 exonuclease resulted in supercoiled mini-seamless vector DNA with only spurious amounts of non-supercoiled (presumably nicked) mini-seamless DNA (lanes 5 and 6; arrow). The substrate DNA digested with NdeI without nuclease treatment is analyzed in lane 3. We determined that about 3 μg of pure supercoiled mini-seamless vector can easily be produced from a 100 ml culture.
Maxi-seamless vector production using variants of attP and attB sites in MG1655-ISC
In another example demonstrating broad utility of strain MG1655-ISC for seamless vector production, we transformed the 13.4 kb substrate plasmid attP4xattH4x (Fig 4A). This recombination substrate carried a 21 bp attB variant (attH4x) and a 241 bp attP variant (attP4x) [13]. Recombination by IntC3 will result in a large supercoiled 10.3 kb seamless vector that carries a human blood clotting factor 8-Ires-hygromycin expression cassette plus a hybrid attL sequence. The second predicted recombination product is a 3.1 kb supercoiled circular DNA that carries bacterial genetic elements plus a hybrid attR sequence.
(A) Map of recombination substrate pattP4xattH4x. The 21 bp attH4x and the 241 bp attP4x sequences are derivatives of the corresponding wild-type attB and attP sites. They flank the 3 kb bacterial backbone as direct repeats. Recombination deletes the bacterial backbone. Relative positions of relevant restriction sites are marked and predicted recombined and un-recombined restriction fragments indicated. (B) A suspension culture (10 ml) was induced for integrase expression by arabinose and further incubation for 70 minutes. Plasmid DNA was purified and analysed by agarose gel electrophoresis. L: 1 kb ladder; Lane 1: substrate vector (before transformation) digested with EcoRV generating two bands of size 7.8 kb and 5.6 kb for linear plasmid; Lane 2: transformed substrate vector (just before induction) digested with EcoRV generating two bands of size 7.8 kb and 5.6 kb for linear unrecombined plasmid; Lane 3: empty; Lane 4: substrate vector (after induction) digested with EcoRV generating four bands of size 7.8 kb and 5.6 kb for linear unrecombined plasmid, 10.3 kb for linear seamless vector and 3.1 kb for linear bacterial backbone; Lane 5: substrate vector (after induction) digested with ClaI generating three bands of size 13.4 kb for linear unrecombined plasmid, 3.1 kb for linear bacterial backbone and 10.3 kb for supercoiled seamless vector; Lane 6: substrate vector (after induction) digested with digested with ClaI and T5 exonuclease generating a supercoiled seamless vector. Note that the additional high molecular weight DNA is most likely the result of intermolecular recombination between seamless vector generating dimers.
Transformed MG1655-ISC cells were induced by arabinose as before, and isolated plasmid DNA was analyzed by agarose gel electrophoresis for recombination products. Results obtained with EcoRV-digested DNA (Fig 4B) revealed that compared to the untransformed substrate DNA (lane 1), the vast majority of plasmid remained unrecombined after transformation to the time point just before induction of IntC3 expression by arabinose (lane 2), hence again demonstrating tight regulation of IntC3 gene expression from the engineered ara operon under the chosen experimental conditions. Induction of IntC3 expression by arabinose produced a majority of the predicted 10.3 kb and 3.1 kb recombination products (EcoRV digest, lane 4). Restriction digestion with ClaI, which cleaves only in the bacterial DNA backbone, resulted in supercoiled 10.3 kb seamless vector, the linearized unrecombined substrate (13.4 kb) and the linearized bacterial backbone (3.1 kb) (lane 5). Moreover, the 10.3 kb seamless vector remained the only product after addition of T5 exonuclease to the ClaI digest (lane 6). In this example, we also observed a possible small fraction of seamless vector dimers, which could be the result of intermolecular recombination between two hybrid attL sites occurring subsequent to intramolecular recombination (lanes 5 and 6). We determined that 60 to 90μg of supercoiled maxi-seamless vector can be produced from a 100ml culture under standard laboratory conditions after phenol/chloroform extraction and ethanol precipitation.
Discussion
We presented here the engineering and application of a novel E. coli strain carrying a tightly regulated gene cassette for the expression of the lambda integrase variant IntC3 inserted into the ara operon. By using a simple cell culture protocol, we demonstrated that IntC3 very efficiently recombined plasmids inside E. coli to yield seamless vectors ranging in size between 500 bp and >10 kb. The employment of restriction enzymes together with the highly active phage T5 exonuclease ultimately resulted in high yields of pure negatively supercoiled seamless vectors.
Novel lambda integrase variants have recently been generated as part of an evolving integrase platform technology that enables site-specific insertion of large transgene cassettes into predetermined endogenous or artificial human genomic att target sequences [13]. Since these genomic targets vary in sequence, it is important to be able to easily generate sufficient seamless target vectors that can accommodate these sequence variations in the corresponding vector-born att sites as recombination partners. The use of our catalytically enhanced IntC3 as recombinase inside E.coli addresses this issue. We demonstrated this by efficient recombination of distinct pairs of att sites on the substrate vectors, e.g. attBPhae2 x attLPhae2 and attH4x x attP4x. This sequence flexibility will expand the lambda integrase att sequence space of potential endogenous target site selection in higher eukaryotes beyond the human genome specifically for seamless payload vector integration. Furthermore, it will be easy to also include specific recombination or cleavage sequences for other genome editing systems on seamless vectors for modular use.
We have shown previously that functional seamless vectors can also be generated by purified integrase in vitro [15]. However, for applications that require large scale production, e.g., as DNA vaccines, producer E. coli strains need to be deployed [22]. Our novel strain MG1655-ISC presented here may thus become a useful means to achieve this objective based on the demonstrated ease in handling and the efficiency of recombination in the absence of leakiness. Furthermore, lambda integrase can recombine the wild-type attP site on an episomal vector with the genomic attB site in E.coli [23] which would lead to bacterial cell toxicity due to incompatible two origins on the genome. Hence, the ability of using variant att site sequences with strain MG1655-ISC may represent an advantage for large scale approaches.
Supporting information
S1 Fig. Sequence of ISC target construct for generating MG1655-ISC.
See text for details.
https://doi.org/10.1371/journal.pone.0270173.s002
(PPTX)
S2 Fig. PCR confirmation of successful targeting events in MG1655-ISC.
(A) Colony PCR was performed with primers ARAC_FWD and INTC3_REV for left junction. PCR amplified products of the expected size 291 bp were detected in all colonies. (B) Colony PCR was performed with primers CAT_FWD and ARAB_REV for right junction. PCR amplified products of the expected size 397 bp were detected in all colonies. (C) Genomic PCR was performed with primers ARAC_FWD and ARAB_REV for full insertion amplification. PCR amplified products of the expected size 2.8 kb were detected in two colonies. L: 100 bp ladder; Lanes 1–4: MG1655/pKD46 colonies electroporated with ISC PCR-amplified construct.
https://doi.org/10.1371/journal.pone.0270173.s003
(PPTX)
S3 Fig. Growth curves of MG1655-ISC and MG1655-IC.
Cells of both strains in stationary phase were inoculated in fresh media and OD600 measured at different time points as indicated. Both strains show very similar exponential growth rates and cell densities in late stationary phase.
https://doi.org/10.1371/journal.pone.0270173.s004
(PPTX)
Acknowledgments
Special thanks to Dr. Michael Berger, University of Münster, Germany, for expert advice in E. coli strain construction. This work was supported through a grant from the National Research Foundation-Competitive Research Programme, Singapore to P.D. (NRF-CRP21-2018-0002). Funding for open access charge: National Research Foundation Competitive Research Programme, Singapore (NRF-CRP21-2018-0002). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
References
- 1. Arevalo-Soliz LM, Hardee CL, Fogg JM, Corman NR, Noorbakhsh C, Zechiedrich L. Improving therapeutic potential of non-viral minimized DNA vectors. Cell Gene Ther Insights. 2020;6(10):1489–505. Epub 2021/05/07. pmid:33953961; PubMed Central PMCID: PMC8095377.
- 2. Hardee CL, Arevalo-Soliz LM, Hornstein BD, Zechiedrich L. Advances in Non-Viral DNA Vectors for Gene Therapy. Genes (Basel). 2017;8(2). Epub 2017/02/18. pmid:28208635; PubMed Central PMCID: PMC5333054.
- 3. Chen ZY, Riu E, He CY, Xu H, Kay MA. Silencing of episomal transgene expression in liver by plasmid bacterial backbone DNA is independent of CpG methylation. Mol Ther. 2008;16(3):548–56. Epub 2008/02/07. pmid:18253155.
- 4. Kay MA. State-of-the-art gene-based therapies: the road ahead. Nat Rev Genet. 2011;12(5):316–28. Epub 2011/04/07. pmid:21468099.
- 5. Mayrhofer P, Blaesen M, Schleef M, Jechlinger W. Minicircle-DNA production by site specific recombination and protein-DNA interaction chromatography. J Gene Med. 2008;10(11):1253–69. Epub 2008/09/04. pmid:18767031.
- 6. Darquet AM, Cameron B, Wils P, Scherman D, Crouzet J. A new DNA vehicle for nonviral gene delivery: supercoiled minicircle. Gene Ther. 1997;4(12):1341–9. Epub 1998/02/24. pmid:9472558.
- 7. Bigger BW, Tolmachov O, Collombet JM, Fragkos M, Palaszewski I, Coutelle C. An araC-controlled bacterial cre expression system to produce DNA minicircle vectors for nuclear and mitochondrial gene therapy. J Biol Chem. 2001;276(25):23018–27. Epub 2001/04/17. pmid:11304530.
- 8. Kristina Nehlsen SB, Juergen Bode. Replicating minicircles: Generation of nonviral episomes for the efficient modification of dividing cells. Gene Therapy and Molecular Biology. 2006;10:233–44.
- 9. Chen ZY, He CY, Ehrhardt A, Kay MA. Minicircle DNA vectors devoid of bacterial DNA result in persistent and high-level transgene expression in vivo. Mol Ther. 2003;8(3):495–500. Epub 2003/08/30. pmid:12946323.
- 10. Jechlinger W, Azimpour Tabrizi C, Lubitz W, Mayrhofer P. Minicircle DNA immobilized in bacterial ghosts: in vivo production of safe non-viral DNA delivery vehicles. J Mol Microbiol Biotechnol. 2004;8(4):222–31. Epub 2005/09/24. pmid:16179799.
- 11. Hou XH, Guo XY, Chen Y, He CY, Chen ZY. Increasing the minicircle DNA purity using an enhanced triplex DNA technology to eliminate DNA contaminants. Mol Ther Methods Clin Dev. 2015;1:14062. Epub 20150114. pmid:26052527; PubMed Central PMCID: PMC4449018.
- 12. Kay MA, He CY, Chen ZY. A robust system for production of minicircle DNA vectors. Nat Biotechnol. 2010;28(12):1287–9. Epub 20101121. pmid:21102455; PubMed Central PMCID: PMC4144359.
- 13. Vijaya Chandra SH, Makhija H, Peter S, Myint Wai CM, Li J, Zhu J, et al. Conservative site-specific and single-copy transgenesis in human LINE-1 elements. Nucleic Acids Res. 2016;44(6):e55. Epub 2015/12/18. pmid:26673710; PubMed Central PMCID: PMC4824084.
- 14. Makhija H, Roy S, Hoon S, Ghadessy FJ, Wong D, Jaiswal R, et al. A novel lambda integrase-mediated seamless vector transgenesis platform for therapeutic protein expression. Nucleic Acids Res. 2018;46(16):e99. Epub 2018/06/13. pmid:29893931; PubMed Central PMCID: PMC6144826.
- 15. Chaudhari N, Rickard AM, Roy S, Droge P, Makhija H. A non-viral genome editing platform for site-specific insertion of large transgenes. Stem Cell Res Ther. 2020;11(1):380. Epub 2020/09/05. pmid:32883366; PubMed Central PMCID: PMC7650303.
- 16. Blattner FR, Plunkett G 3rd, Bloch CA, Perna NT, Burland V, Riley M, et al. The complete genome sequence of Escherichia coli K-12. Science. 1997;277(5331):1453–62. Epub 1997/09/05. pmid:9278503.
- 17. Guyer MS, Reed RR, Steitz JA, Low KB. Identification of a sex-factor-affinity site in E. coli as gamma delta. Cold Spring Harb Symp Quant Biol. 1981;45 Pt 1:135–40. Epub 1981/01/01. pmid:6271456.
- 18.
Bachmann BBJ. Derivations and genotypes of some mutant derivatives of Escherichia coli K-12. In Neidhardt (ed), Escherichia coli and Salmonella, 2nd ed ASM Press, Washington, DC. 1996:2460–88.
- 19. Corona T, Bao Q, Christ N, Schwartz T, Li J, Droge P. Activation of site-specific DNA integration in human cells by a single chain integration host factor. Nucleic Acids Res. 2003;31(17):5140–8. Epub 2003/08/22. pmid:12930965; PubMed Central PMCID: PMC212810.
- 20. Thomason L, Court DL, Bubunenko M, Costantino N, Wilson H, Datta S, et al. Recombineering: genetic engineering in bacteria using homologous recombination. Curr Protoc Mol Biol. 2007;Chapter 1:Unit 1 16. Epub 2008/02/12. pmid:18265390.
- 21. 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(12):6640–5. Epub 2000/06/01. pmid:10829079; PubMed Central PMCID: PMC18686.
- 22. Gaspar V, de Melo-Diogo D, Costa E, Moreira A, Queiroz J, Pichon C, et al. Minicircle DNA vectors for gene therapy: advances and applications. Expert Opin Biol Ther. 2015;15(3):353–79. Epub 2014/12/30. pmid:25539147.
- 23. Siau JW, Chee S, Makhija H, Wai CM, Chandra SH, Peter S, et al. Directed evolution of lambda integrase activity and specificity by genetic derepression. Protein Eng Des Sel. 2015;28(7):211–20. Epub 2015/03/20. pmid:25787692.